15 skills found
sayantann11 / All Classification Templetes For MLClassification - Machine Learning This is ‘Classification’ tutorial which is a part of the Machine Learning course offered by Simplilearn. We will learn Classification algorithms, types of classification algorithms, support vector machines(SVM), Naive Bayes, Decision Tree and Random Forest Classifier in this tutorial. Objectives Let us look at some of the objectives covered under this section of Machine Learning tutorial. Define Classification and list its algorithms Describe Logistic Regression and Sigmoid Probability Explain K-Nearest Neighbors and KNN classification Understand Support Vector Machines, Polynomial Kernel, and Kernel Trick Analyze Kernel Support Vector Machines with an example Implement the Naïve Bayes Classifier Demonstrate Decision Tree Classifier Describe Random Forest Classifier Classification: Meaning Classification is a type of supervised learning. It specifies the class to which data elements belong to and is best used when the output has finite and discrete values. It predicts a class for an input variable as well. There are 2 types of Classification: Binomial Multi-Class Classification: Use Cases Some of the key areas where classification cases are being used: To find whether an email received is a spam or ham To identify customer segments To find if a bank loan is granted To identify if a kid will pass or fail in an examination Classification: Example Social media sentiment analysis has two potential outcomes, positive or negative, as displayed by the chart given below. https://www.simplilearn.com/ice9/free_resources_article_thumb/classification-example-machine-learning.JPG This chart shows the classification of the Iris flower dataset into its three sub-species indicated by codes 0, 1, and 2. https://www.simplilearn.com/ice9/free_resources_article_thumb/iris-flower-dataset-graph.JPG The test set dots represent the assignment of new test data points to one class or the other based on the trained classifier model. Types of Classification Algorithms Let’s have a quick look into the types of Classification Algorithm below. Linear Models Logistic Regression Support Vector Machines Nonlinear models K-nearest Neighbors (KNN) Kernel Support Vector Machines (SVM) Naïve Bayes Decision Tree Classification Random Forest Classification Logistic Regression: Meaning Let us understand the Logistic Regression model below. This refers to a regression model that is used for classification. This method is widely used for binary classification problems. It can also be extended to multi-class classification problems. Here, the dependent variable is categorical: y ϵ {0, 1} A binary dependent variable can have only two values, like 0 or 1, win or lose, pass or fail, healthy or sick, etc In this case, you model the probability distribution of output y as 1 or 0. This is called the sigmoid probability (σ). If σ(θ Tx) > 0.5, set y = 1, else set y = 0 Unlike Linear Regression (and its Normal Equation solution), there is no closed form solution for finding optimal weights of Logistic Regression. Instead, you must solve this with maximum likelihood estimation (a probability model to detect the maximum likelihood of something happening). It can be used to calculate the probability of a given outcome in a binary model, like the probability of being classified as sick or passing an exam. https://www.simplilearn.com/ice9/free_resources_article_thumb/logistic-regression-example-graph.JPG Sigmoid Probability The probability in the logistic regression is often represented by the Sigmoid function (also called the logistic function or the S-curve): https://www.simplilearn.com/ice9/free_resources_article_thumb/sigmoid-function-machine-learning.JPG In this equation, t represents data values * the number of hours studied and S(t) represents the probability of passing the exam. Assume sigmoid function: https://www.simplilearn.com/ice9/free_resources_article_thumb/sigmoid-probability-machine-learning.JPG g(z) tends toward 1 as z -> infinity , and g(z) tends toward 0 as z -> infinity K-nearest Neighbors (KNN) K-nearest Neighbors algorithm is used to assign a data point to clusters based on similarity measurement. It uses a supervised method for classification. The steps to writing a k-means algorithm are as given below: https://www.simplilearn.com/ice9/free_resources_article_thumb/knn-distribution-graph-machine-learning.JPG Choose the number of k and a distance metric. (k = 5 is common) Find k-nearest neighbors of the sample that you want to classify Assign the class label by majority vote. KNN Classification A new input point is classified in the category such that it has the most number of neighbors from that category. For example: https://www.simplilearn.com/ice9/free_resources_article_thumb/knn-classification-machine-learning.JPG Classify a patient as high risk or low risk. Mark email as spam or ham. Keen on learning about Classification Algorithms in Machine Learning? Click here! Support Vector Machine (SVM) Let us understand Support Vector Machine (SVM) in detail below. SVMs are classification algorithms used to assign data to various classes. They involve detecting hyperplanes which segregate data into classes. SVMs are very versatile and are also capable of performing linear or nonlinear classification, regression, and outlier detection. Once ideal hyperplanes are discovered, new data points can be easily classified. https://www.simplilearn.com/ice9/free_resources_article_thumb/support-vector-machines-graph-machine-learning.JPG The optimization objective is to find “maximum margin hyperplane” that is farthest from the closest points in the two classes (these points are called support vectors). In the given figure, the middle line represents the hyperplane. SVM Example Let’s look at this image below and have an idea about SVM in general. Hyperplanes with larger margins have lower generalization error. The positive and negative hyperplanes are represented by: https://www.simplilearn.com/ice9/free_resources_article_thumb/positive-negative-hyperplanes-machine-learning.JPG Classification of any new input sample xtest : If w0 + wTxtest > 1, the sample xtest is said to be in the class toward the right of the positive hyperplane. If w0 + wTxtest < -1, the sample xtest is said to be in the class toward the left of the negative hyperplane. When you subtract the two equations, you get: https://www.simplilearn.com/ice9/free_resources_article_thumb/equation-subtraction-machine-learning.JPG Length of vector w is (L2 norm length): https://www.simplilearn.com/ice9/free_resources_article_thumb/length-of-vector-machine-learning.JPG You normalize with the length of w to arrive at: https://www.simplilearn.com/ice9/free_resources_article_thumb/normalize-equation-machine-learning.JPG SVM: Hard Margin Classification Given below are some points to understand Hard Margin Classification. The left side of equation SVM-1 given above can be interpreted as the distance between the positive (+ve) and negative (-ve) hyperplanes; in other words, it is the margin that can be maximized. Hence the objective of the function is to maximize with the constraint that the samples are classified correctly, which is represented as : https://www.simplilearn.com/ice9/free_resources_article_thumb/hard-margin-classification-machine-learning.JPG This means that you are minimizing ‖w‖. This also means that all positive samples are on one side of the positive hyperplane and all negative samples are on the other side of the negative hyperplane. This can be written concisely as : https://www.simplilearn.com/ice9/free_resources_article_thumb/hard-margin-classification-formula.JPG Minimizing ‖w‖ is the same as minimizing. This figure is better as it is differentiable even at w = 0. The approach listed above is called “hard margin linear SVM classifier.” SVM: Soft Margin Classification Given below are some points to understand Soft Margin Classification. To allow for linear constraints to be relaxed for nonlinearly separable data, a slack variable is introduced. (i) measures how much ith instance is allowed to violate the margin. The slack variable is simply added to the linear constraints. https://www.simplilearn.com/ice9/free_resources_article_thumb/soft-margin-calculation-machine-learning.JPG Subject to the above constraints, the new objective to be minimized becomes: https://www.simplilearn.com/ice9/free_resources_article_thumb/soft-margin-calculation-formula.JPG You have two conflicting objectives now—minimizing slack variable to reduce margin violations and minimizing to increase the margin. The hyperparameter C allows us to define this trade-off. Large values of C correspond to larger error penalties (so smaller margins), whereas smaller values of C allow for higher misclassification errors and larger margins. https://www.simplilearn.com/ice9/free_resources_article_thumb/machine-learning-certification-video-preview.jpg SVM: Regularization The concept of C is the reverse of regularization. Higher C means lower regularization, which increases bias and lowers the variance (causing overfitting). https://www.simplilearn.com/ice9/free_resources_article_thumb/concept-of-c-graph-machine-learning.JPG IRIS Data Set The Iris dataset contains measurements of 150 IRIS flowers from three different species: Setosa Versicolor Viriginica Each row represents one sample. Flower measurements in centimeters are stored as columns. These are called features. IRIS Data Set: SVM Let’s train an SVM model using sci-kit-learn for the Iris dataset: https://www.simplilearn.com/ice9/free_resources_article_thumb/svm-model-graph-machine-learning.JPG Nonlinear SVM Classification There are two ways to solve nonlinear SVMs: by adding polynomial features by adding similarity features Polynomial features can be added to datasets; in some cases, this can create a linearly separable dataset. https://www.simplilearn.com/ice9/free_resources_article_thumb/nonlinear-classification-svm-machine-learning.JPG In the figure on the left, there is only 1 feature x1. This dataset is not linearly separable. If you add x2 = (x1)2 (figure on the right), the data becomes linearly separable. Polynomial Kernel In sci-kit-learn, one can use a Pipeline class for creating polynomial features. Classification results for the Moons dataset are shown in the figure. https://www.simplilearn.com/ice9/free_resources_article_thumb/polynomial-kernel-machine-learning.JPG Polynomial Kernel with Kernel Trick Let us look at the image below and understand Kernel Trick in detail. https://www.simplilearn.com/ice9/free_resources_article_thumb/polynomial-kernel-with-kernel-trick.JPG For large dimensional datasets, adding too many polynomial features can slow down the model. You can apply a kernel trick with the effect of polynomial features without actually adding them. The code is shown (SVC class) below trains an SVM classifier using a 3rd-degree polynomial kernel but with a kernel trick. https://www.simplilearn.com/ice9/free_resources_article_thumb/polynomial-kernel-equation-machine-learning.JPG The hyperparameter coefθ controls the influence of high-degree polynomials. Kernel SVM Let us understand in detail about Kernel SVM. Kernel SVMs are used for classification of nonlinear data. In the chart, nonlinear data is projected into a higher dimensional space via a mapping function where it becomes linearly separable. https://www.simplilearn.com/ice9/free_resources_article_thumb/kernel-svm-machine-learning.JPG In the higher dimension, a linear separating hyperplane can be derived and used for classification. A reverse projection of the higher dimension back to original feature space takes it back to nonlinear shape. As mentioned previously, SVMs can be kernelized to solve nonlinear classification problems. You can create a sample dataset for XOR gate (nonlinear problem) from NumPy. 100 samples will be assigned the class sample 1, and 100 samples will be assigned the class label -1. https://www.simplilearn.com/ice9/free_resources_article_thumb/kernel-svm-graph-machine-learning.JPG As you can see, this data is not linearly separable. https://www.simplilearn.com/ice9/free_resources_article_thumb/kernel-svm-non-separable.JPG You now use the kernel trick to classify XOR dataset created earlier. https://www.simplilearn.com/ice9/free_resources_article_thumb/kernel-svm-xor-machine-learning.JPG Naïve Bayes Classifier What is Naive Bayes Classifier? Have you ever wondered how your mail provider implements spam filtering or how online news channels perform news text classification or even how companies perform sentiment analysis of their audience on social media? All of this and more are done through a machine learning algorithm called Naive Bayes Classifier. Naive Bayes Named after Thomas Bayes from the 1700s who first coined this in the Western literature. Naive Bayes classifier works on the principle of conditional probability as given by the Bayes theorem. Advantages of Naive Bayes Classifier Listed below are six benefits of Naive Bayes Classifier. Very simple and easy to implement Needs less training data Handles both continuous and discrete data Highly scalable with the number of predictors and data points As it is fast, it can be used in real-time predictions Not sensitive to irrelevant features Bayes Theorem We will understand Bayes Theorem in detail from the points mentioned below. According to the Bayes model, the conditional probability P(Y|X) can be calculated as: P(Y|X) = P(X|Y)P(Y) / P(X) This means you have to estimate a very large number of P(X|Y) probabilities for a relatively small vector space X. For example, for a Boolean Y and 30 possible Boolean attributes in the X vector, you will have to estimate 3 billion probabilities P(X|Y). To make it practical, a Naïve Bayes classifier is used, which assumes conditional independence of P(X) to each other, with a given value of Y. This reduces the number of probability estimates to 2*30=60 in the above example. Naïve Bayes Classifier for SMS Spam Detection Consider a labeled SMS database having 5574 messages. It has messages as given below: https://www.simplilearn.com/ice9/free_resources_article_thumb/naive-bayes-spam-machine-learning.JPG Each message is marked as spam or ham in the data set. Let’s train a model with Naïve Bayes algorithm to detect spam from ham. The message lengths and their frequency (in the training dataset) are as shown below: https://www.simplilearn.com/ice9/free_resources_article_thumb/naive-bayes-spam-spam-detection.JPG Analyze the logic you use to train an algorithm to detect spam: Split each message into individual words/tokens (bag of words). Lemmatize the data (each word takes its base form, like “walking” or “walked” is replaced with “walk”). Convert data to vectors using scikit-learn module CountVectorizer. Run TFIDF to remove common words like “is,” “are,” “and.” Now apply scikit-learn module for Naïve Bayes MultinomialNB to get the Spam Detector. This spam detector can then be used to classify a random new message as spam or ham. Next, the accuracy of the spam detector is checked using the Confusion Matrix. For the SMS spam example above, the confusion matrix is shown on the right. Accuracy Rate = Correct / Total = (4827 + 592)/5574 = 97.21% Error Rate = Wrong / Total = (155 + 0)/5574 = 2.78% https://www.simplilearn.com/ice9/free_resources_article_thumb/confusion-matrix-machine-learning.JPG Although confusion Matrix is useful, some more precise metrics are provided by Precision and Recall. https://www.simplilearn.com/ice9/free_resources_article_thumb/precision-recall-matrix-machine-learning.JPG Precision refers to the accuracy of positive predictions. https://www.simplilearn.com/ice9/free_resources_article_thumb/precision-formula-machine-learning.JPG Recall refers to the ratio of positive instances that are correctly detected by the classifier (also known as True positive rate or TPR). https://www.simplilearn.com/ice9/free_resources_article_thumb/recall-formula-machine-learning.JPG Precision/Recall Trade-off To detect age-appropriate videos for kids, you need high precision (low recall) to ensure that only safe videos make the cut (even though a few safe videos may be left out). The high recall is needed (low precision is acceptable) in-store surveillance to catch shoplifters; a few false alarms are acceptable, but all shoplifters must be caught. Learn about Naive Bayes in detail. Click here! Decision Tree Classifier Some aspects of the Decision Tree Classifier mentioned below are. Decision Trees (DT) can be used both for classification and regression. The advantage of decision trees is that they require very little data preparation. They do not require feature scaling or centering at all. They are also the fundamental components of Random Forests, one of the most powerful ML algorithms. Unlike Random Forests and Neural Networks (which do black-box modeling), Decision Trees are white box models, which means that inner workings of these models are clearly understood. In the case of classification, the data is segregated based on a series of questions. Any new data point is assigned to the selected leaf node. https://www.simplilearn.com/ice9/free_resources_article_thumb/decision-tree-classifier-machine-learning.JPG Start at the tree root and split the data on the feature using the decision algorithm, resulting in the largest information gain (IG). This splitting procedure is then repeated in an iterative process at each child node until the leaves are pure. This means that the samples at each node belonging to the same class. In practice, you can set a limit on the depth of the tree to prevent overfitting. The purity is compromised here as the final leaves may still have some impurity. The figure shows the classification of the Iris dataset. https://www.simplilearn.com/ice9/free_resources_article_thumb/decision-tree-classifier-graph.JPG IRIS Decision Tree Let’s build a Decision Tree using scikit-learn for the Iris flower dataset and also visualize it using export_graphviz API. https://www.simplilearn.com/ice9/free_resources_article_thumb/iris-decision-tree-machine-learning.JPG The output of export_graphviz can be converted into png format: https://www.simplilearn.com/ice9/free_resources_article_thumb/iris-decision-tree-output.JPG Sample attribute stands for the number of training instances the node applies to. Value attribute stands for the number of training instances of each class the node applies to. Gini impurity measures the node’s impurity. A node is “pure” (gini=0) if all training instances it applies to belong to the same class. https://www.simplilearn.com/ice9/free_resources_article_thumb/impurity-formula-machine-learning.JPG For example, for Versicolor (green color node), the Gini is 1-(0/54)2 -(49/54)2 -(5/54) 2 ≈ 0.168 https://www.simplilearn.com/ice9/free_resources_article_thumb/iris-decision-tree-sample.JPG Decision Boundaries Let us learn to create decision boundaries below. For the first node (depth 0), the solid line splits the data (Iris-Setosa on left). Gini is 0 for Setosa node, so no further split is possible. The second node (depth 1) splits the data into Versicolor and Virginica. If max_depth were set as 3, a third split would happen (vertical dotted line). https://www.simplilearn.com/ice9/free_resources_article_thumb/decision-tree-boundaries.JPG For a sample with petal length 5 cm and petal width 1.5 cm, the tree traverses to depth 2 left node, so the probability predictions for this sample are 0% for Iris-Setosa (0/54), 90.7% for Iris-Versicolor (49/54), and 9.3% for Iris-Virginica (5/54) CART Training Algorithm Scikit-learn uses Classification and Regression Trees (CART) algorithm to train Decision Trees. CART algorithm: Split the data into two subsets using a single feature k and threshold tk (example, petal length < “2.45 cm”). This is done recursively for each node. k and tk are chosen such that they produce the purest subsets (weighted by their size). The objective is to minimize the cost function as given below: https://www.simplilearn.com/ice9/free_resources_article_thumb/cart-training-algorithm-machine-learning.JPG The algorithm stops executing if one of the following situations occurs: max_depth is reached No further splits are found for each node Other hyperparameters may be used to stop the tree: min_samples_split min_samples_leaf min_weight_fraction_leaf max_leaf_nodes Gini Impurity or Entropy Entropy is one more measure of impurity and can be used in place of Gini. https://www.simplilearn.com/ice9/free_resources_article_thumb/gini-impurity-entrophy.JPG It is a degree of uncertainty, and Information Gain is the reduction that occurs in entropy as one traverses down the tree. Entropy is zero for a DT node when the node contains instances of only one class. Entropy for depth 2 left node in the example given above is: https://www.simplilearn.com/ice9/free_resources_article_thumb/entrophy-for-depth-2.JPG Gini and Entropy both lead to similar trees. DT: Regularization The following figure shows two decision trees on the moons dataset. https://www.simplilearn.com/ice9/free_resources_article_thumb/dt-regularization-machine-learning.JPG The decision tree on the right is restricted by min_samples_leaf = 4. The model on the left is overfitting, while the model on the right generalizes better. Random Forest Classifier Let us have an understanding of Random Forest Classifier below. A random forest can be considered an ensemble of decision trees (Ensemble learning). Random Forest algorithm: Draw a random bootstrap sample of size n (randomly choose n samples from the training set). Grow a decision tree from the bootstrap sample. At each node, randomly select d features. Split the node using the feature that provides the best split according to the objective function, for instance by maximizing the information gain. Repeat the steps 1 to 2 k times. (k is the number of trees you want to create, using a subset of samples) Aggregate the prediction by each tree for a new data point to assign the class label by majority vote (pick the group selected by the most number of trees and assign new data point to that group). Random Forests are opaque, which means it is difficult to visualize their inner workings. https://www.simplilearn.com/ice9/free_resources_article_thumb/random-forest-classifier-graph.JPG However, the advantages outweigh their limitations since you do not have to worry about hyperparameters except k, which stands for the number of decision trees to be created from a subset of samples. RF is quite robust to noise from the individual decision trees. Hence, you need not prune individual decision trees. The larger the number of decision trees, the more accurate the Random Forest prediction is. (This, however, comes with higher computation cost). Key Takeaways Let us quickly run through what we have learned so far in this Classification tutorial. Classification algorithms are supervised learning methods to split data into classes. They can work on Linear Data as well as Nonlinear Data. Logistic Regression can classify data based on weighted parameters and sigmoid conversion to calculate the probability of classes. K-nearest Neighbors (KNN) algorithm uses similar features to classify data. Support Vector Machines (SVMs) classify data by detecting the maximum margin hyperplane between data classes. Naïve Bayes, a simplified Bayes Model, can help classify data using conditional probability models. Decision Trees are powerful classifiers and use tree splitting logic until pure or somewhat pure leaf node classes are attained. Random Forests apply Ensemble Learning to Decision Trees for more accurate classification predictions. Conclusion This completes ‘Classification’ tutorial. In the next tutorial, we will learn 'Unsupervised Learning with Clustering.'
Aryia-Behroziuan / NeuronsAn ANN is a model based on a collection of connected units or nodes called "artificial neurons", which loosely model the neurons in a biological brain. Each connection, like the synapses in a biological brain, can transmit information, a "signal", from one artificial neuron to another. An artificial neuron that receives a signal can process it and then signal additional artificial neurons connected to it. In common ANN implementations, the signal at a connection between artificial neurons is a real number, and the output of each artificial neuron is computed by some non-linear function of the sum of its inputs. The connections between artificial neurons are called "edges". Artificial neurons and edges typically have a weight that adjusts as learning proceeds. The weight increases or decreases the strength of the signal at a connection. Artificial neurons may have a threshold such that the signal is only sent if the aggregate signal crosses that threshold. Typically, artificial neurons are aggregated into layers. Different layers may perform different kinds of transformations on their inputs. Signals travel from the first layer (the input layer) to the last layer (the output layer), possibly after traversing the layers multiple times. The original goal of the ANN approach was to solve problems in the same way that a human brain would. However, over time, attention moved to performing specific tasks, leading to deviations from biology. Artificial neural networks have been used on a variety of tasks, including computer vision, speech recognition, machine translation, social network filtering, playing board and video games and medical diagnosis. Deep learning consists of multiple hidden layers in an artificial neural network. This approach tries to model the way the human brain processes light and sound into vision and hearing. Some successful applications of deep learning are computer vision and speech recognition.[68] Decision trees Main article: Decision tree learning Decision tree learning uses a decision tree as a predictive model to go from observations about an item (represented in the branches) to conclusions about the item's target value (represented in the leaves). It is one of the predictive modeling approaches used in statistics, data mining, and machine learning. Tree models where the target variable can take a discrete set of values are called classification trees; in these tree structures, leaves represent class labels and branches represent conjunctions of features that lead to those class labels. Decision trees where the target variable can take continuous values (typically real numbers) are called regression trees. In decision analysis, a decision tree can be used to visually and explicitly represent decisions and decision making. In data mining, a decision tree describes data, but the resulting classification tree can be an input for decision making. Support vector machines Main article: Support vector machines Support vector machines (SVMs), also known as support vector networks, are a set of related supervised learning methods used for classification and regression. Given a set of training examples, each marked as belonging to one of two categories, an SVM training algorithm builds a model that predicts whether a new example falls into one category or the other.[69] An SVM training algorithm is a non-probabilistic, binary, linear classifier, although methods such as Platt scaling exist to use SVM in a probabilistic classification setting. In addition to performing linear classification, SVMs can efficiently perform a non-linear classification using what is called the kernel trick, implicitly mapping their inputs into high-dimensional feature spaces. Illustration of linear regression on a data set. Regression analysis Main article: Regression analysis Regression analysis encompasses a large variety of statistical methods to estimate the relationship between input variables and their associated features. Its most common form is linear regression, where a single line is drawn to best fit the given data according to a mathematical criterion such as ordinary least squares. The latter is often extended by regularization (mathematics) methods to mitigate overfitting and bias, as in ridge regression. When dealing with non-linear problems, go-to models include polynomial regression (for example, used for trendline fitting in Microsoft Excel[70]), logistic regression (often used in statistical classification) or even kernel regression, which introduces non-linearity by taking advantage of the kernel trick to implicitly map input variables to higher-dimensional space. Bayesian networks Main article: Bayesian network A simple Bayesian network. Rain influences whether the sprinkler is activated, and both rain and the sprinkler influence whether the grass is wet. A Bayesian network, belief network, or directed acyclic graphical model is a probabilistic graphical model that represents a set of random variables and their conditional independence with a directed acyclic graph (DAG). For example, a Bayesian network could represent the probabilistic relationships between diseases and symptoms. Given symptoms, the network can be used to compute the probabilities of the presence of various diseases. Efficient algorithms exist that perform inference and learning. Bayesian networks that model sequences of variables, like speech signals or protein sequences, are called dynamic Bayesian networks. Generalizations of Bayesian networks that can represent and solve decision problems under uncertainty are called influence diagrams. Genetic algorithms Main article: Genetic algorithm A genetic algorithm (GA) is a search algorithm and heuristic technique that mimics the process of natural selection, using methods such as mutation and crossover to generate new genotypes in the hope of finding good solutions to a given problem. In machine learning, genetic algorithms were used in the 1980s and 1990s.[71][72] Conversely, machine learning techniques have been used to improve the performance of genetic and evolutionary algorithms.[73] Training models Usually, machine learning models require a lot of data in order for them to perform well. Usually, when training a machine learning model, one needs to collect a large, representative sample of data from a training set. Data from the training set can be as varied as a corpus of text, a collection of images, and data collected from individual users of a service. Overfitting is something to watch out for when training a machine learning model. Federated learning Main article: Federated learning Federated learning is an adapted form of distributed artificial intelligence to training machine learning models that decentralizes the training process, allowing for users' privacy to be maintained by not needing to send their data to a centralized server. This also increases efficiency by decentralizing the training process to many devices. For example, Gboard uses federated machine learning to train search query prediction models on users' mobile phones without having to send individual searches back to Google.[74] Applications There are many applications for machine learning, including: Agriculture Anatomy Adaptive websites Affective computing Banking Bioinformatics Brain–machine interfaces Cheminformatics Citizen science Computer networks Computer vision Credit-card fraud detection Data quality DNA sequence classification Economics Financial market analysis[75] General game playing Handwriting recognition Information retrieval Insurance Internet fraud detection Linguistics Machine learning control Machine perception Machine translation Marketing Medical diagnosis Natural language processing Natural language understanding Online advertising Optimization Recommender systems Robot locomotion Search engines Sentiment analysis Sequence mining Software engineering Speech recognition Structural health monitoring Syntactic pattern recognition Telecommunication Theorem proving Time series forecasting User behavior analytics In 2006, the media-services provider Netflix held the first "Netflix Prize" competition to find a program to better predict user preferences and improve the accuracy of its existing Cinematch movie recommendation algorithm by at least 10%. A joint team made up of researchers from AT&T Labs-Research in collaboration with the teams Big Chaos and Pragmatic Theory built an ensemble model to win the Grand Prize in 2009 for $1 million.[76] Shortly after the prize was awarded, Netflix realized that viewers' ratings were not the best indicators of their viewing patterns ("everything is a recommendation") and they changed their recommendation engine accordingly.[77] In 2010 The Wall Street Journal wrote about the firm Rebellion Research and their use of machine learning to predict the financial crisis.[78] In 2012, co-founder of Sun Microsystems, Vinod Khosla, predicted that 80% of medical doctors' jobs would be lost in the next two decades to automated machine learning medical diagnostic software.[79] In 2014, it was reported that a machine learning algorithm had been applied in the field of art history to study fine art paintings and that it may have revealed previously unrecognized influences among artists.[80] In 2019 Springer Nature published the first research book created using machine learning.[81] Limitations Although machine learning has been transformative in some fields, machine-learning programs often fail to deliver expected results.[82][83][84] Reasons for this are numerous: lack of (suitable) data, lack of access to the data, data bias, privacy problems, badly chosen tasks and algorithms, wrong tools and people, lack of resources, and evaluation problems.[85] In 2018, a self-driving car from Uber failed to detect a pedestrian, who was killed after a collision.[86] Attempts to use machine learning in healthcare with the IBM Watson system failed to deliver even after years of time and billions of dollars invested.[87][88] Bias Main article: Algorithmic bias Machine learning approaches in particular can suffer from different data biases. A machine learning system trained on current customers only may not be able to predict the needs of new customer groups that are not represented in the training data. When trained on man-made data, machine learning is likely to pick up the same constitutional and unconscious biases already present in society.[89] Language models learned from data have been shown to contain human-like biases.[90][91] Machine learning systems used for criminal risk assessment have been found to be biased against black people.[92][93] In 2015, Google photos would often tag black people as gorillas,[94] and in 2018 this still was not well resolved, but Google reportedly was still using the workaround to remove all gorillas from the training data, and thus was not able to recognize real gorillas at all.[95] Similar issues with recognizing non-white people have been found in many other systems.[96] In 2016, Microsoft tested a chatbot that learned from Twitter, and it quickly picked up racist and sexist language.[97] Because of such challenges, the effective use of machine learning may take longer to be adopted in other domains.[98] Concern for fairness in machine learning, that is, reducing bias in machine learning and propelling its use for human good is increasingly expressed by artificial intelligence scientists, including Fei-Fei Li, who reminds engineers that "There’s nothing artificial about AI...It’s inspired by people, it’s created by people, and—most importantly—it impacts people. It is a powerful tool we are only just beginning to understand, and that is a profound responsibility.”[99] Model assessments Classification of machine learning models can be validated by accuracy estimation techniques like the holdout method, which splits the data in a training and test set (conventionally 2/3 training set and 1/3 test set designation) and evaluates the performance of the training model on the test set. In comparison, the K-fold-cross-validation method randomly partitions the data into K subsets and then K experiments are performed each respectively considering 1 subset for evaluation and the remaining K-1 subsets for training the model. In addition to the holdout and cross-validation methods, bootstrap, which samples n instances with replacement from the dataset, can be used to assess model accuracy.[100] In addition to overall accuracy, investigators frequently report sensitivity and specificity meaning True Positive Rate (TPR) and True Negative Rate (TNR) respectively. Similarly, investigators sometimes report the false positive rate (FPR) as well as the false negative rate (FNR). However, these rates are ratios that fail to reveal their numerators and denominators. The total operating characteristic (TOC) is an effective method to express a model's diagnostic ability. TOC shows the numerators and denominators of the previously mentioned rates, thus TOC provides more information than the commonly used receiver operating characteristic (ROC) and ROC's associated area under the curve (AUC).[101] Ethics Machine learning poses a host of ethical questions. Systems which are trained on datasets collected with biases may exhibit these biases upon use (algorithmic bias), thus digitizing cultural prejudices.[102] For example, using job hiring data from a firm with racist hiring policies may lead to a machine learning system duplicating the bias by scoring job applicants against similarity to previous successful applicants.[103][104] Responsible collection of data and documentation of algorithmic rules used by a system thus is a critical part of machine learning. Because human languages contain biases, machines trained on language corpora will necessarily also learn these biases.[105][106] Other forms of ethical challenges, not related to personal biases, are more seen in health care. There are concerns among health care professionals that these systems might not be designed in the public's interest but as income-generating machines. This is especially true in the United States where there is a long-standing ethical dilemma of improving health care, but also increasing profits. For example, the algorithms could be designed to provide patients with unnecessary tests or medication in which the algorithm's proprietary owners hold stakes. There is huge potential for machine learning in health care to provide professionals a great tool to diagnose, medicate, and even plan recovery paths for patients, but this will not happen until the personal biases mentioned previously, and these "greed" biases are addressed.[107] Hardware Since the 2010s, advances in both machine learning algorithms and computer hardware have led to more efficient methods for training deep neural networks (a particular narrow subdomain of machine learning) that contain many layers of non-linear hidden units.[108] By 2019, graphic processing units (GPUs), often with AI-specific enhancements, had displaced CPUs as the dominant method of training large-scale commercial cloud AI.[109] OpenAI estimated the hardware compute used in the largest deep learning projects from AlexNet (2012) to AlphaZero (2017), and found a 300,000-fold increase in the amount of compute required, with a doubling-time trendline of 3.4 months.[110][111] Software Software suites containing a variety of machine learning algorithms include the following: Free and open-source so
SOYJUN / FTP Implement Based On UDPThe aim of this assignment is to have you do UDP socket client / server programming with a focus on two broad aspects : Setting up the exchange between the client and server in a secure way despite the lack of a formal connection (as in TCP) between the two, so that ‘outsider’ UDP datagrams (broadcast, multicast, unicast - fortuitously or maliciously) cannot intrude on the communication. Introducing application-layer protocol data-transmission reliability, flow control and congestion control in the client and server using TCP-like ARQ sliding window mechanisms. The second item above is much more of a challenge to implement than the first, though neither is particularly trivial. But they are not tightly interdependent; each can be worked on separately at first and then integrated together at a later stage. Apart from the material in Chapters 8, 14 & 22 (especially Sections 22.5 - 22.7), and the experience you gained from the preceding assignment, you will also need to refer to the following : ioctl function (Chapter 17). get_ifi_info function (Section 17.6, Chapter 17). This function will be used by the server code to discover its node’s network interfaces so that it can bind all its interface IP addresses (see Section 22.6). ‘Race’ conditions (Section 20.5, Chapter 20) You also need a thorough understanding of how the TCP protocol implements reliable data transfer, flow control and congestion control. Chapters 17- 24 of TCP/IP Illustrated, Volume 1 by W. Richard Stevens gives a good overview of TCP. Though somewhat dated for some things (it was published in 1994), it remains, overall, a good basic reference. Overview This assignment asks you to implement a primitive file transfer protocol for Unix platforms, based on UDP, and with TCP-like reliability added to the transfer operation using timeouts and sliding-window mechanisms, and implementing flow and congestion control. The server is a concurrent server which can handle multiple clients simultaneously. A client gives the server the name of a file. The server forks off a child which reads directly from the file and transfers the contents over to the client using UDP datagrams. The client prints out the file contents as they come in, in order, with nothing missing and with no duplication of content, directly on to stdout (via the receiver sliding window, of course, but with no other intermediate buffering). The file to be transferred can be of arbitrary length, but its contents are always straightforward ascii text. As an aside let me mention that assuming the file contents ascii is not as restrictive as it sounds. We can always pretend, for example, that binary files are base64 encoded (“ASCII armor”). A real file transfer protocol would, of course, have to worry about transferring files between heterogeneous platforms with different file structure conventions and semantics. The sender would first have to transform the file into a platform-independent, protocol-defined, format (using, say, ASN.1, or some such standard), and the receiver would have to transform the received file into its platform’s native file format. This kind of thing can be fairly time consuming, and is certainly very tedious, to implement, with little educational value - it is not part of this assignment. Arguments for the server You should provide the server with an input file server.in from which it reads the following information, in the order shown, one item per line : Well-known port number for server. Maximum sending sliding-window size (in datagram units). You will not be handing in your server.in file. We shall create our own when we come to test your code. So it is important that you stick strictly to the file name and content conventions specified above. The same applies to the client.in input file below. Arguments for the client The client is to be provided with an input file client.in from which it reads the following information, in the order shown, one item per line : IP address of server (not the hostname). Well-known port number of server. filename to be transferred. Receiving sliding-window size (in datagram units). Random generator seed value. Probability p of datagram loss. This should be a real number in the range [ 0.0 , 1.0 ] (value 0.0 means no loss occurs; value 1.0 means all datagrams all lost). The mean µ, in milliseconds, for an exponential distribution controlling the rate at which the client reads received datagram payloads from its receive buffer. Operation Server starts up and reads its arguments from file server.in. As we shall see, when a client communicates with the server, the server will want to know what IP address that client is using to identify the server (i.e. , the destination IP address in the incoming datagram). Normally, this can be done relatively straightforwardly using the IP_RECVDESTADDR socket option, and picking up the information using the ancillary data (‘control information’) capability of the recvmsg function. Unfortunately, Solaris 2.10 does not support the IP_RECVDESTADDR option (nor, incidentally, does it support the msg_flags option in msghdr - see p.390). This considerably complicates things. In the absence of IP_RECVDESTADDR, what the server has to do as part of its initialization phase is to bind each IP address it has (and, simultaneously, its well-known port number, which it has read in from server.in) to a separate UDP socket. The code in Section 22.6, which uses the get_ifi_info function, shows you how to do that. However, there are important differences between that code and the version you want to implement. The code of Section 22.6 binds the IP addresses and forks off a child for each address that is bound to. We do not want to do that. Instead you should have an array of socket descriptors. For each IP address, create a new socket and bind the address (and well-known port number) to the socket without forking off child processes. Creating child processes comes later, when clients arrive. The code of Section 22.6 also attempts to bind broadcast addresses. We do not want to do this. It binds a wildcard IP address, which we certainly do not want to do either. We should bind strictly only unicast addresses (including the loopback address). The get_ifi_info function (which the code in Section 22.6 uses) has to be modified so that it also gets the network masks for the IP addresses of the node, and adds these to the information stored in the linked list of ifi_info structures (see Figure 17.5, p.471) it produces. As you go binding each IP address to a distinct socket, it will be useful for later processing to build your own array of structures, where a structure element records the following information for each socket : sockfd IP address bound to the socket network mask for the IP address subnet address (obtained by doing a bit-wise and between the IP address and its network mask) Report, in a ReadMe file which you hand in with your code, on the modifications you had to introduce to ensure that only unicast addresses are bound, and on your implementation of the array of structures described above. You should print out on stdout, with an appropriate message and appropriately formatted in dotted decimal notation, the IP address, network mask, and subnet address for each socket in your array of structures (you do not need to print the sockfd). The server now uses select to monitor the sockets it has created for incoming datagrams. When it returns from select, it must use recvfrom or recvmsg to read the incoming datagram (see 6. below). When a client starts, it first reads its arguments from the file client.in. The client checks if the server host is ‘local’ to its (extended) Ethernet. If so, all its communication to the server is to occur as MSG_DONTROUTE (or SO_DONTROUTE socket option). It determines if the server host is ‘local’ as follows. The first thing the client should do is to use the modified get_ifi_info function to obtain all of its IP addresses and associated network masks. Print out on stdout, in dotted decimal notation and with an appropriate message, the IP addresses and network masks obtained. In the following, IPserver designates the IP address the client will use to identify the server, and IPclient designates the IP address the client will choose to identify itself. The client checks whether the server is on the same host. If so, it should use the loopback address 127.0.0.1 for the server (i.e. , IPserver = 127.0.0.1). IPclient should also be set to the loopback address. Otherwise it proceeds as follows: IPserver is set to the IP address for the server in the client.in file. Given IPserver and the (unicast) IP addresses and network masks for the client returned by get_ifi_info in the linked list of ifi_info structures, you should be able to figure out if the server node is ‘local’ or not. This will be discussed in class; but let me just remind you here that you should use ‘longest prefix matching’ where applicable. If there are multiple client addresses, and the server host is ‘local’, the client chooses an IP address for itself, IPclient, which matches up as ‘local’ according to your examination above. If the server host is not ‘local’, then IPclient can be chosen arbitrarily. Print out on stdout the results of your examination, as to whether the server host is ‘local’ or not, as well as the IPclient and IPserver addresses selected. Note that this manner of determining whether the server is local or not is somewhat clumsy and ‘over-engineered’, and, as such, should be viewed more in the nature of a pedagogical exercise. Ideally, we would like to look up the server IP address(es) in the routing table (see Section 18.3). This requires that a routing socket be created, for which we need superuser privilege. Alternatively, we might want to dump out the routing table, using the sysctl function for example (see Section 18.4), and examine it directly. Unfortunately, Solaris 2.10 does not support sysctl. Furthermore, note that there is a slight problem with the address 130.245.1.123/24 assigned to compserv3 (see rightmost column of file hosts, and note that this particular compserv3 address “overlaps” with the 130.245.1.x/28 addresses in that same column assigned to compserv1, compserv2 & comserv4). In particular, if the client is running on compserv3 and the server on any of the other three compservs, and if that server node is also being identified to the client by its /28 (rather than its /24) address, then the client will get a “false positive” when it tests as to whether the server node is local or not. In other words, the client will deem the server node to be local, whereas in fact it should not be considered local. Because of this, it is perhaps best simply not to use compserv3 to run the client (but it is o.k. to use it to run the server). Finally, using MSG_DONTROUTE where possible would seem to gain us efficiency, in as much as the kernel does not need to consult the routing table for every datagram sent. But, in fact, that is not so. Recall that one effect of connect with UDP sockets is that routing information is obtained by the kernel at the time the connect is issued. That information is cached and used for subsequent sends from the connected socket (see p.255). The client now creates a UDP socket and calls bind on IPclient, with 0 as the port number. This will cause the kernel to bind an ephemeral port to the socket. After the bind, use the getsockname function (Section 4.10) to obtain IPclient and the ephemeral port number that has been assigned to the socket, and print that information out on stdout, with an appropriate message and appropriately formatted. The client connects its socket to IPserver and the well-known port number of the server. After the connect, use the getpeername function (Section 4.10) to obtain IPserver and the well-known port number of the server, and print that information out on stdout, with an appropriate message and appropriately formatted. The client sends a datagram to the server giving the filename for the transfer. This send needs to be backed up by a timeout in case the datagram is lost. Note that the incoming datagram from the client will be delivered to the server at the socket to which the destination IP address that the datagram is carrying has been bound. Thus, the server can obtain that address (it is, of course, IPserver) and thereby achieve what IP_RECVDESTADDR would have given us had it been available. Furthermore, the server process can obtain the IP address (this will, of course, be IPclient) and ephemeral port number of the client through the recvfrom or recvmsg functions. The server forks off a child process to handle the client. The server parent process goes back to the select to listen for new clients. Hereafter, and unless otherwise stated, whenever we refer to the ‘server’, we mean the server child process handling the client’s file transfer, not the server parent process. Typically, the first thing the server child would be expected to do is to close all sockets it ‘inherits’ from its parent. However, this is not the case with us. The server child does indeed close the sockets it inherited, but not the socket on which the client request arrived. It leaves that socket open for now. Call this socket the ‘listening’ socket. The server (child) then checks if the client host is local to its (extended) Ethernet. If so, all its communication to the client is to occur as MSG_DONTROUTE (or SO_DONTROUTE socket option). If IPserver (obtained in 5. above) is the loopback address, then we are done. Otherwise, the server has to proceed with the following step. Use the array of structures you built in 1. above, together with the addresses IPserver and IPclient to determine if the client is ‘local’. Print out on stdout the results of your examination, as to whether the client host is ‘local’ or not. The server (child) creates a UDP socket to handle file transfer to the client. Call this socket the ‘connection’ socket. It binds the socket to IPserver, with port number 0 so that its kernel assigns an ephemeral port. After the bind, use the getsockname function (Section 4.10) to obtain IPserver and the ephemeral port number that has been assigned to the socket, and print that information out on stdout, with an appropriate message and appropriately formatted. The server then connects this ‘connection’ socket to the client’s IPclient and ephemeral port number. The server now sends the client a datagram, in which it passes it the ephemeral port number of its ‘connection’ socket as the data payload of the datagram. This datagram is sent using the ‘listening’ socket inherited from its parent, otherwise the client (whose socket is connected to the server’s ‘listening’ socket at the latter’s well-known port number) will reject it. This datagram must be backed up by the ARQ mechanism, and retransmitted in the event of loss. Note that if this datagram is indeed lost, the client might well time out and retransmit its original request message (the one carrying the file name). In this event, you must somehow ensure that the parent server does not mistake this retransmitted request for a new client coming in, and spawn off yet another child to handle it. How do you do that? It is potentially more involved than it might seem. I will be discussing this in class, as well as ‘race’ conditions that could potentially arise, depending on how you code the mechanisms I present. When the client receives the datagram carrying the ephemeral port number of the server’s ‘connection’ socket, it reconnects its socket to the server’s ‘connection’ socket, using IPserver and the ephemeral port number received in the datagram (see p.254). It now uses this reconnected socket to send the server an acknowledgment. Note that this implies that, in the event of the server timing out, it should retransmit two copies of its ‘ephemeral port number’ message, one on its ‘listening’ socket and the other on its ‘connection’ socket (why?). When the server receives the acknowledgment, it closes the ‘listening’ socket it inherited from its parent. The server can now commence the file transfer through its ‘connection’ socket. The net effect of all these binds and connects at server and client is that no ‘outsider’ UDP datagram (broadcast, multicast, unicast - fortuitously or maliciously) can now intrude on the communication between server and client. Starting with the first datagram sent out, the client behaves as follows. Whenever a datagram arrives, or an ACK is about to be sent out (or, indeed, the initial datagram to the server giving the filename for the transfer), the client uses some random number generator function random() (initialized by the client.in argument value seed) to decide with probability p (another client.in argument value) if the datagram or ACK should be discarded by way of simulating transmission loss across the network. (I will briefly discuss in class how you do this.) Adding reliability to UDP The mechanisms you are to implement are based on TCP Reno. These include : Reliable data transmission using ARQ sliding-windows, with Fast Retransmit. Flow control via receiver window advertisements. Congestion control that implements : SlowStart Congestion Avoidance (‘Additive-Increase/Multiplicative Decrease’ – AIMD) Fast Recovery (but without the window-inflation aspect of Fast Recovery) Only some, and by no means all, of the details for these are covered below. The rest will be presented in class, especially those concerning flow control and TCP Reno’s congestion control mechanisms in general : Slow Start, Congestion Avoidance, Fast Retransmit and Fast Recovery. Implement a timeout mechanism on the sender (server) side. This is available to you from Stevens, Section 22.5 . Note, however, that you will need to modify the basic driving mechanism of Figure 22.7 appropriately since the situation at the sender side is not a repetitive cycle of send-receive, but rather a straightforward progression of send-send-send-send- . . . . . . . . . . . Also, modify the RTT and RTO mechanisms of Section 22.5 as specified below. I will be discussing the details of these modifications and the reasons for them in class. Modify function rtt_stop (Fig. 22.13) so that it uses integer arithmetic rather than floating point. This will entail your also having to modify some of the variable and function parameter declarations throughout Section 22.5 from float to int, as appropriate. In the unprrt.h header file (Fig. 22.10) set : RTT_RXTMIN to 1000 msec. (1 sec. instead of the current value 3 sec.) RTT_RXTMAX to 3000 msec. (3 sec. instead of the current value 60 sec.) RTT_MAXNREXMT to 12 (instead of the current value 3) In function rtt_timeout (Fig. 22.14), after doubling the RTO in line 86, pass its value through the function rtt_minmax of Fig. 22.11 (somewhat along the lines of what is done in line 77 of rtt_stop, Fig. 22.13). Finally, note that with the modification to integer calculation of the smoothed RTT and its variation, and given the small RTT values you will experience on the cs / sbpub network, these calculations should probably now be done on a millisecond or even microsecond scale (rather than in seconds, as is the case with Stevens’ code). Otherwise, small measured RTTs could show up as 0 on a scale of seconds, yielding a negative result when we subtract the smoothed RTT from the measured RTT (line 72 of rtt_stop, Fig. 22.13). Report the details of your modifications to the code of Section 22.5 in the ReadMe file which you hand in with your code. We need to have a sender sliding window mechanism for the retransmission of lost datagrams; and a receiver sliding window in order to ensure correct sequencing of received file contents, and some measure of flow control. You should implement something based on TCP Reno’s mechanisms, with cumulative acknowledgments, receiver window advertisements, and a congestion control mechanism I will explain in detail in class. For a reference on TCP’s mechanisms generally, see W. Richard Stevens, TCP/IP Illustrated, Volume 1 , especially Sections 20.2 - 20.4 of Chapter 20 , and Sections 21.1 - 21.8 of Chapter 21 . Bear in mind that our sequence numbers should count datagrams, not bytes as in TCP. Remember that the sender and receiver window sizes have to be set according to the argument values in client.in and server.in, respectively. Whenever the sender window becomes full and so ‘locks’, the server should print out a message to that effect on stdout. Similarly, whenever the receiver window ‘locks’, the client should print out a message on stdout. Be aware of the potential for deadlock when the receiver window ‘locks’. This situation is handled by having the receiver process send a duplicate ACK which acts as a window update when its window opens again (see Figure 20.3 and the discussion about it in TCP/IP Illustrated). However, this is not enough, because ACKs are not backed up by a timeout mechanism in the event they are lost. So we will also need to implement a persist timer driving window probes in the sender process (see Sections 22.1 & 22.2 in Chapter 22 of TCP/IP Illustrated). Note that you do not have to worry about the Silly Window Syndrome discussed in Section 22.3 of TCP/IP Illustrated since the receiver process consumes ‘full sized’ 512-byte messages from the receiver buffer (see 3. below). Report on the details of the ARQ mechanism you implemented in the ReadMe file you hand in. Indeed, you should report on all the TCP mechanisms you implemented in the ReadMe file, both the ones discussed here, and the ones I will be discussing in class. Make your datagram payload a fixed 512 bytes, inclusive of the file transfer protocol header (which must, at the very least, carry: the sequence number of the datagram; ACKs; and advertised window notifications). The client reads the file contents in its receive buffer and prints them out on stdout using a separate thread. This thread sits in a repetitive loop till all the file contents have been printed out, doing the following. It samples from an exponential distribution with mean µ milliseconds (read from the client.in file), sleeps for that number of milliseconds; wakes up to read and print all in-order file contents available in the receive buffer at that point; samples again from the exponential distribution; sleeps; and so on. The formula -1 × µ × ln( random( ) ) , where ln is the natural logarithm, yields variates from an exponential distribution with mean µ, based on the uniformly-distributed variates over ( 0 , 1 ) returned by random(). Note that you will need to implement some sort of mutual exclusion/semaphore mechanism on the client side so that the thread that sleeps and wakes up to consume from the receive buffer is not updating the state variables of the buffer at the same time as the main thread reading from the socket and depositing into the buffer is doing the same. Furthermore, we need to ensure that the main thread does not effectively monopolize the semaphore (and thus lock out for prolonged periods of time) the sleeping thread when the latter wakes up. See the textbook, Section 26.7, ‘Mutexes: Mutual Exclusion’, pp.697-701. You might also find Section 26.8, ‘Condition Variables’, pp.701-705, useful. You will need to devise some way by which the sender can notify the receiver when it has sent the last datagram of the file transfer, without the receiver mistaking that EOF marker as part of the file contents. (Also, note that the last data segment could be a “short” segment of less than 512 bytes – your client needs to be able to handle this correctly somehow.) When the sender receives an ACK for the last datagram of the transfer, the (child) server terminates. The parent server has to take care of cleaning up zombie children. Note that if we want a clean closing, the client process cannot simply terminate when the receiver ACKs the last datagram. This ACK could be lost, which would leave the (child) server process ‘hanging’, timing out, and retransmitting the last datagram. TCP attempts to deal with this problem by means of the TIME_WAIT state. You should have your receiver process behave similarly, sticking around in something akin to a TIME_WAIT state in case in case it needs to retransmit the ACK. In the ReadMe file you hand in, report on how you dealt with the issues raised here: sender notifying receiver of the last datagram, clean closing, and so on. Output Some of the output required from your program has been described in the section Operation above. I expect you to provide further output – clear, well-structured, well-laid-out, concise but sufficient and helpful – in the client and server windows by means of which we can trace the correct evolution of your TCP’s behaviour in all its intricacies : information (e.g., sequence number) on datagrams and acks sent and dropped, window advertisements, datagram retransmissions (and why : dup acks or RTO); entering/exiting Slow Start and Congestion Avoidance, ssthresh and cwnd values; sender and receiver windows locking/unlocking; etc., etc. . . . . The onus is on you to convince us that the TCP mechanisms you implemented are working correctly. Too many students do not put sufficient thought, creative imagination, time or effort into this. It is not the TA’s nor my responsibility to sit staring at an essentially blank screen, trying to summon up our paranormal psychology skills to figure out if your TCP implementation is really working correctly in all its very intricate aspects, simply because the transferred file seems to be printing o.k. in the client window. Nor is it our responsibility to strain our eyes and our patience wading through a mountain of obscure, ill-structured, hyper-messy, debugging-style output because, for example, your effort-conserving concept of what is ‘suitable’ is to dump your debugging output on us, relevant, irrelevant, and everything in between.
Rynkll696 / HHimport pyttsx3 import speech_recognition as sr import datetime from datetime import date import calendar import time import math import wikipedia import webbrowser import os import smtplib import winsound import pyautogui import cv2 from pygame import mixer from tkinter import * import tkinter.messagebox as message from sqlite3 import * conn = connect("voice_assistant_asked_questions.db") conn.execute("CREATE TABLE IF NOT EXISTS `voicedata`(id INTEGER PRIMARY KEY AUTOINCREMENT,command VARCHAR(201))") conn.execute("CREATE TABLE IF NOT EXISTS `review`(id INTEGER PRIMARY KEY AUTOINCREMENT, review VARCHAR(50), type_of_review VARCHAR(50))") conn.execute("CREATE TABLE IF NOT EXISTS `emoji`(id INTEGER PRIMARY KEY AUTOINCREMENT,emoji VARCHAR(201))") global query engine = pyttsx3.init('sapi5') voices = engine.getProperty('voices') engine.setProperty('voice', voices[0].id) def speak(audio): engine.say(audio) engine.runAndWait() def wishMe(): hour = int(datetime.datetime.now().hour) if hour >= 0 and hour<12: speak("Good Morning!") elif hour >= 12 and hour < 18: speak("Good Afternoon!") else: speak("Good Evening!") speak("I am voice assistant Akshu2020 Sir. Please tell me how may I help you.") def takeCommand(): global query r = sr.Recognizer() with sr.Microphone() as source: print("Listening...") r.pause_threshold = 0.9 audio = r.listen(source) try: print("Recognizing...") query = r.recognize_google(audio,language='en-in') print(f"User said: {query}\n") except Exception as e: #print(e) print("Say that again please...") #speak('Say that again please...') return "None" return query def calculator(): global query try: if 'add' in query or 'edi' in query: speak('Enter a number') a = float(input("Enter a number:")) speak('Enter another number to add') b = float(input("Enter another number to add:")) c = a+b print(f"{a} + {b} = {c}") speak(f'The addition of {a} and {b} is {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'sub' in query: speak('Enter a number') a = float(input("Enter a number:")) speak('Enter another number to subtract') b = float(input("Enter another number to subtract:")) c = a-b print(f"{a} - {b} = {c}") speak(f'The subtraction of {a} and {b} is {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'mod' in query: speak('Enter a number') a = float(input("Enter a number:")) speak('Enter another number') b = float(input("Enter another number:")) c = a%b print(f"{a} % {b} = {c}") speak(f'The modular division of {a} and {b} is equal to {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'div' in query: speak('Enter a number as dividend') a = float(input("Enter a number:")) speak('Enter another number as divisor') b = float(input("Enter another number as divisor:")) c = a/b print(f"{a} / {b} = {c}") speak(f'{a} divided by {b} is equal to {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'multi' in query: speak('Enter a number') a = float(input("Enter a number:")) speak('Enter another number to multiply') b = float(input("Enter another number to multiply:")) c = a*b print(f"{a} x {b} = {c}") speak(f'The multiplication of {a} and {b} is {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'square root' in query: speak('Enter a number to find its sqare root') a = float(input("Enter a number:")) c = a**(1/2) print(f"Square root of {a} = {c}") speak(f'Square root of {a} is {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'square' in query: speak('Enter a number to find its sqare') a = float(input("Enter a number:")) c = a**2 print(f"{a} x {a} = {c}") speak(f'Square of {a} is {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'cube root' in query: speak('Enter a number to find its cube root') a = float(input("Enter a number:")) c = a**(1/3) print(f"Cube root of {a} = {c}") speak(f'Cube root of {a} is {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'cube' in query: speak('Enter a number to find its sqare') a = float(input("Enter a number:")) c = a**3 print(f"{a} x {a} x {a} = {c}") speak(f'Cube of {a} is {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'fact' in query: try: n = int(input('Enter the number whose factorial you want to find:')) fact = 1 for i in range(1,n+1): fact = fact*i print(f"{n}! = {fact}") speak(f'{n} factorial is equal to {fact}. Your answer is {fact}.') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') except Exception as e: #print(e) speak('I unable to calculate its factorial.') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'power' in query or 'raise' in query: speak('Enter a number whose power you want to raised') a = float(input("Enter a number whose power to be raised :")) speak(f'Enter a raised power to {a}') b = float(input(f"Enter a raised power to {a}:")) c = a**b print(f"{a} ^ {b} = {c}") speak(f'{a} raise to the power {b} = {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'percent' in query: speak('Enter a number whose percentage you want to calculate') a = float(input("Enter a number whose percentage you want to calculate :")) speak(f'How many percent of {a} you want to calculate?') b = float(input(f"Enter how many percentage of {a} you want to calculate:")) c = (a*b)/100 print(f"{b} % of {a} is {c}") speak(f'{b} percent of {a} is {c}. Your answer is {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'interest' in query: speak('Enter the principal value or amount') p = float(input("Enter the principal value (P):")) speak('Enter the rate of interest per year') r = float(input("Enter the rate of interest per year (%):")) speak('Enter the time in months') t = int(input("Enter the time (in months):")) interest = (p*r*t)/1200 sint = round(interest) fv = round(p + interest) print(f"Interest = {interest}") print(f"The total amount accured, principal plus interest, from simple interest on a principal of {p} at a rate of {r}% per year for {t} months is {p + interest}.") speak(f'interest is {sint}. The total amount accured, principal plus interest, from simple interest on a principal of {p} at a rate of {r}% per year for {t} months is {fv}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'si' in query: speak('Enter the angle in degree to find its sine value') a = float(input("Enter the angle:")) b = a * 3.14/180 c = math.sin(b) speak('Here is your answer.') print(f"sin({a}) = {c}") speak(f'sin({a}) = {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'cos' in query: speak('Enter the angle in degree to find its cosine value') a = float(input("Enter the angle:")) b = a * 3.14/180 c = math.cos(b) speak('Here is your answer.') print(f"cos({a}) = {c}") speak(f'cos({a}) = {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'cot' in query or 'court' in query: try: speak('Enter the angle in degree to find its cotangent value') a = float(input("Enter the angle:")) b = a * 3.14/180 c = 1/math.tan(b) speak('Here is your answer.') print(f"cot({a}) = {c}") speak(f'cot({a}) = {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') except Exception as e: print("infinity") speak('Answer is infinity') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'tan' in query or '10' in query: speak('Enter the angle in degree to find its tangent value') a = float(input("Enter the angle:")) b = a * 3.14/180 c = math.tan(b) speak('Here is your answer.') print(f"tan({a}) = {c}") speak(f'tan({a}) = {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'cosec' in query: try: speak('Enter the angle in degree to find its cosecant value') a = float(input("Enter the angle:")) b = a * 3.14/180 c =1/ math.sin(b) speak('Here is your answer.') print(f"cosec({a}) = {c}") speak(f'cosec({a}) = {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') except Exception as e: print('Infinity') speak('Answer is infinity') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'caus' in query: try: speak('Enter the angle in degree to find its cosecant value') a = float(input("Enter the angle:")) b = a * 3.14/180 c =1/ math.sin(b) speak('Here is your answer.') print(f"cosec({a}) = {c}") speak(f'cosec({a}) = {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') except Exception as e: print('Infinity') speak('Answer is infinity') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') elif 'sec' in query: try: speak('Enter the angle in degree to find its secant value') a = int(input("Enter the angle:")) b = a * 3.14/180 c = 1/math.cos(b) speak('Here is your answer.') print(f"sec({a}) = {c}") speak(f'sec({a}) = {c}') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') except Exception as e: print('Infinity') speak('Answer is infinity') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') except Exception as e: speak('I unable to do this calculation.') speak('Do you want to do another calculation?') query = takeCommand().lower() if 'y' in query: speak('ok which calculation you want to do?') query = takeCommand().lower() calculator() else: speak('ok') def callback(r,c): global player if player == 'X' and states[r][c] == 0 and stop_game == False: b[r][c].configure(text='X',fg='blue', bg='white') states[r][c] = 'X' player = 'O' if player == 'O' and states[r][c] == 0 and stop_game == False: b[r][c].configure(text='O',fg='red', bg='yellow') states[r][c] = 'O' player = 'X' check_for_winner() def check_for_winner(): global stop_game global root for i in range(3): if states[i][0] == states[i][1]== states[i][2]!=0: b[i][0].config(bg='grey') b[i][1].config(bg='grey') b[i][2].config(bg='grey') stop_game = True root.destroy() for i in range(3): if states[0][i] == states[1][i] == states[2][i]!= 0: b[0][i].config(bg='grey') b[1][i].config(bg='grey') b[2][i].config(bg='grey') stop_game = True root.destroy() if states[0][0] == states[1][1]== states[2][2]!= 0: b[0][0].config(bg='grey') b[1][1].config(bg='grey') b[2][2].config(bg='grey') stop_game = True root.destroy() if states[2][0] == states[1][1] == states[0][2]!= 0: b[2][0].config(bg='grey') b[1][1].config(bg='grey') b[0][2].config(bg='grey') stop_game = True root.destroy() def sendEmail(to,content): server = smtplib.SMTP('smtp.gmail.com', 587) server.ehlo() server.starttls() server.login('xyz123@gmail.com','password') server.sendmail('xyz123@gmail.com',to,content) server.close() def brightness(): try: query = takeCommand().lower() if '25' in query: pyautogui.moveTo(1880,1050) pyautogui.click() time.sleep(1) pyautogui.moveTo(1610,960) pyautogui.click() pyautogui.moveTo(1880,1050) pyautogui.click() speak('If you again want to change brihtness, say, change brightness') elif '50' in query: pyautogui.moveTo(1880,1050) pyautogui.click() time.sleep(1) pyautogui.moveTo(1684,960) pyautogui.click() pyautogui.moveTo(1880,1050) pyautogui.click() speak('If you again want to change brihtness, say, change brightness') elif '75' in query: pyautogui.moveTo(1880,1050) pyautogui.click() time.sleep(1) pyautogui.moveTo(1758,960) pyautogui.click() pyautogui.moveTo(1880,1050) pyautogui.click() speak('If you again want to change brihtness, say, change brightness') elif '100' in query or 'full' in query: pyautogui.moveTo(1880,1050) pyautogui.click() time.sleep(1) pyautogui.moveTo(1835,960) pyautogui.click() pyautogui.moveTo(1880,1050) pyautogui.click() speak('If you again want to change brihtness, say, change brightness') else: speak('Please select 25, 50, 75 or 100....... Say again.') brightness() except exception as e: #print(e) speak('Something went wrong') def close_window(): try: if 'y' in query: pyautogui.moveTo(1885,10) pyautogui.click() else: speak('ok') pyautogui.moveTo(1000,500) except exception as e: #print(e) speak('error') def whatsapp(): query = takeCommand().lower() if 'y' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('whatsapp') time.sleep(2) pyautogui.press('enter') time.sleep(2) pyautogui.moveTo(100,140) pyautogui.click() speak('To whom you want to send message,.....just write the name here in 5 seconds') time.sleep(7) pyautogui.moveTo(120,300) pyautogui.click() time.sleep(1) pyautogui.moveTo(800,990) pyautogui.click() speak('Say the message,....or if you want to send anything else,...say send document, or say send emoji') query = takeCommand() if ('sent' in query or 'send' in query) and 'document' in query: pyautogui.moveTo(660,990) pyautogui.click() time.sleep(1) pyautogui.moveTo(660,740) pyautogui.click() speak('please select the document within 10 seconds') time.sleep(12) speak('Should I send this document?') query = takeCommand().lower() if 'y' in query and 'no' not in query: speak('sending the document......') pyautogui.press('enter') speak('Do you want to send message again to anyone?') whatsapp() elif ('remove' in query or 'cancel' in query or 'delete' in query or 'clear' in query) and ('document' in query or 'message' in query or 'it' in query or 'emoji' in query or 'select' in query): pyautogui.doubleClick(x=800, y=990) pyautogui.press('backspace') speak('Do you want to send message again to anyone?') whatsapp() else: speak('ok') elif ('sent' in query or 'send' in query) and 'emoji' in query: pyautogui.moveTo(620,990) pyautogui.click() pyautogui.moveTo(670,990) pyautogui.click() pyautogui.moveTo(650,580) pyautogui.click() speak('please select the emoji within 10 seconds') time.sleep(11) speak('Should I send this emoji?') query = takeCommand().lower() if 'y' in query and 'no' not in query: speak('Sending the emoji......') pyautogui.press('enter') speak('Do you want to send message again to anyone?') whatsapp() elif ('remove' in query or 'cancel' in query or 'delete' in query or 'clear' in query) and ('message' in query or 'it' in query or 'emoji' in query or 'select' in query): pyautogui.doublClick(x=800, y=990) speak('Do you want to send message again to anyone?') whatsapp() else: speak('ok') else: pyautogui.write(f'{query}') speak('Should I send this message?') query = takeCommand().lower() if 'y' in query and 'no' not in query: speak('sending the message......') pyautogui.press('enter') speak('Do you want to send message again to anyone?') whatsapp() elif ('remove' in query or 'cancel' in query or 'delete' in query or 'clear' in query) and ('message' in query or 'it' in query or 'select' in query): pyautogui.doubleClick(x=800, y=990) pyautogui.press('backspace') speak('Do you want to send message again to anyone?') whatsapp() else: speak('ok') else: speak('ok') def alarm(): root = Tk() root.title('Akshu2020 Alarm-Clock') speak('Please enter the time in the format hour, minutes and seconds. When the alarm should rang?') speak('Please enter the time greater than the current time') def setalarm(): alarmtime = f"{hrs.get()}:{mins.get()}:{secs.get()}" print(alarmtime) if(alarmtime!="::"): alarmclock(alarmtime) else: speak('You have not entered the time.') def alarmclock(alarmtime): while True: time.sleep(1) time_now=datetime.datetime.now().strftime("%H:%M:%S") print(time_now) if time_now == alarmtime: Wakeup=Label(root, font = ('arial', 20, 'bold'), text="Wake up! Wake up! Wake up",bg="DodgerBlue2",fg="white").grid(row=6,columnspan=3) speak("Wake up, Wake up") print("Wake up!") mixer.init() mixer.music.load(r'C:\Users\Admin\Music\Playlists\wake-up-will-you-446.mp3') mixer.music.play() break speak('you can click on close icon to close the alarm window.') hrs=StringVar() mins=StringVar() secs=StringVar() greet=Label(root, font = ('arial', 20, 'bold'),text="Take a short nap!").grid(row=1,columnspan=3) hrbtn=Entry(root,textvariable=hrs,width=5,font =('arial', 20, 'bold')) hrbtn.grid(row=2,column=1) minbtn=Entry(root,textvariable=mins, width=5,font = ('arial', 20, 'bold')).grid(row=2,column=2) secbtn=Entry(root,textvariable=secs, width=5,font = ('arial', 20, 'bold')).grid(row=2,column=3) setbtn=Button(root,text="set alarm",command=setalarm,bg="DodgerBlue2", fg="white",font = ('arial', 20, 'bold')).grid(row=4,columnspan=3) timeleft = Label(root,font=('arial', 20, 'bold')) timeleft.grid() mainloop() def select1(): global vs global root3 global type_of_review if vs.get() == 1: message.showinfo(" ","Thank you for your review!!") review = "Very Satisfied" type_of_review = "Positive" root3.destroy() elif vs.get() == 2: message.showinfo(" ","Thank you for your review!!") review = "Satisfied" type_of_review = "Positive" root3.destroy() elif vs.get() == 3: message.showinfo(" ","Thank you for your review!!!!") review = "Neither Satisfied Nor Dissatisfied" type_of_review = "Neutral" root3.destroy() elif vs.get() == 4: message.showinfo(" ","Thank you for your review!!") review = "Dissatisfied" type_of_review = "Negative" root3.destroy() elif vs.get() == 5: message.showinfo(" ","Thank you for your review!!") review = "Very Dissatisfied" type_of_review = "Negative" root3.destroy() elif vs.get() == 6: message.showinfo(" "," Ok ") review = "I do not want to give review" type_of_review = "No review" root3.destroy() try: conn.execute(f"INSERT INTO `review`(review,type_of_review) VALUES('{review}', '{type_of_review}')") conn.commit() except Exception as e: pass def select_review(): global root3 global vs global type_of_review root3 = Tk() root3.title("Select an option") vs = IntVar() string = "Are you satisfied with my performance?" msgbox = Message(root3,text=string) msgbox.config(bg="lightgreen",font = "(20)") msgbox.grid(row=0,column=0) rs1=Radiobutton(root3,text="Very Satisfied",font="(20)",value=1,variable=vs).grid(row=1,column=0,sticky=W) rs2=Radiobutton(root3,text="Satisfied",font="(20)",value=2,variable=vs).grid(row=2,column=0,sticky=W) rs3=Radiobutton(root3,text="Neither Satisfied Nor Dissatisfied",font="(20)",value=3,variable=vs).grid(row=3,column=0,sticky=W) rs4=Radiobutton(root3,text="Dissatisfied",font="(20)",value=4,variable=vs).grid(row=4,column=0,sticky=W) rs5=Radiobutton(root3,text="Very Dissatisfied",font="(20)",value=5,variable=vs).grid(row=5,column=0,sticky=W) rs6=Radiobutton(root3,text="I don't want to give review",font="(20)",value=6,variable=vs).grid(row=6,column=0,sticky=W) bs = Button(root3,text="Submit",font="(20)",activebackground="yellow",activeforeground="green",command=select1) bs.grid(row=7,columnspan=2) root3.mainloop() while True : query = takeCommand().lower() # logic for executing tasks based on query if 'wikipedia' in query: speak('Searching wikipedia...') query = query.replace("wikipedia","") results = wikipedia.summary(query, sentences=2) speak("According to Wikipedia") print(results) speak(results) elif 'translat' in query or ('let' in query and 'translat' in query and 'open' in query): webbrowser.open('https://translate.google.co.in') time.sleep(10) elif 'open map' in query or ('let' in query and 'map' in query and 'open' in query): webbrowser.open('https://www.google.com/maps') time.sleep(10) elif ('open' in query and 'youtube' in query) or ('let' in query and 'youtube' in query and 'open' in query): webbrowser.open('https://www.youtube.com') time.sleep(10) elif 'chrome' in query: webbrowser.open('https://www.chrome.com') time.sleep(10) elif 'weather' in query: webbrowser.open('https://www.yahoo.com/news/weather') time.sleep(3) speak('Click on, change location, and enter the city , whose whether conditions you want to know.') time.sleep(10) elif 'google map' in query: webbrowser.open('https://www.google.com/maps') time.sleep(10) elif ('open' in query and 'google' in query) or ('let' in query and 'google' in query and 'open' in query): webbrowser.open('google.com') time.sleep(10) elif ('open' in query and 'stack' in query and 'overflow' in query) or ('let' in query and 'stack' in query and 'overflow' in query and 'open' in query): webbrowser.open('stackoverflow.com') time.sleep(10) elif 'open v i' in query or 'open vi' in query or 'open vierp' in query or ('open' in query and ('r p' in query or 'rp' in query)): webbrowser.open('https://www.vierp.in/login/erplogin') time.sleep(10) elif 'news' in query: webbrowser.open('https://www.bbc.com/news/world') time.sleep(10) elif 'online shop' in query or (('can' in query or 'want' in query or 'do' in query or 'could' in query) and 'shop' in query) or('let' in query and 'shop' in query): speak('From which online shopping website, you want to shop? Amazon, flipkart, snapdeal or naaptol?') query = takeCommand().lower() if 'amazon' in query: webbrowser.open('https://www.amazon.com') time.sleep(10) elif 'flip' in query: webbrowser.open('https://www.flipkart.com') time.sleep(10) elif 'snap' in query: webbrowser.open('https://www.snapdeal.com') time.sleep(10) elif 'na' in query: webbrowser.open('https://www.naaptol.com') time.sleep(10) else: speak('Sorry sir, you have to search in browser as his shopping website is not reachable for me.') elif ('online' in query and ('game' in query or 'gaming' in query)): webbrowser.open('https://www.agame.com/games') time.sleep(10) elif 'dictionary' in query: webbrowser.open('https://www.dictionary.com') time.sleep(3) speak('Enter the word, in the search bar of the dictionary, whose defination or synonyms you want to know') time.sleep(3) elif ('identif' in query and 'emoji' in query) or ('sentiment' in query and ('analysis' in query or 'identif' in query)): speak('Please enter only one emoji at a time.') emoji = input('enter emoji here: ') if '😀' in emoji or '😃' in emoji or '😄' in emoji or '😁' in emoji or '🙂' in emoji or '😊' in emoji or '☺️' in emoji or '😇' in emoji or '🥲' in emoji: speak('happy') print('Happy') elif '😝' in emoji or '😆' in emoji or '😂' in emoji or '🤣' in emoji: speak('Laughing') print('Laughing') elif '😡' in emoji or '😠' in emoji or '🤬' in emoji: speak('Angry') print('Angry') elif '🤫' in emoji: speak('Keep quite') print('Keep quite') elif '😷' in emoji: speak('face with mask') print('Face with mask') elif '🥳' in emoji: speak('party') print('party') elif '😢' in emoji or '😥' in emoji or '😓' in emoji or '😰' in emoji or '☹️' in emoji or '🙁' in emoji or '😟' in emoji or '😔' in emoji or '😞️' in emoji: speak('Sad') print('Sad') elif '😭' in emoji: speak('Crying') print('Crying') elif '😋' in emoji: speak('Tasty') print('Tasty') elif '🤨' in emoji: speak('Doubt') print('Doubt') elif '😴' in emoji: speak('Sleeping') print('Sleeping') elif '🥱' in emoji: speak('feeling sleepy') print('feeling sleepy') elif '😍' in emoji or '🥰' in emoji or '😘' in emoji: speak('Lovely') print('Lovely') elif '😱' in emoji: speak('Horrible') print('Horrible') elif '🎂' in emoji: speak('Cake') print('Cake') elif '🍫' in emoji: speak('Cadbury') print('Cadbury') elif '🇮🇳' in emoji: speak('Indian national flag,.....Teeranga') print('Indian national flag - Tiranga') elif '💐' in emoji: speak('Bouquet') print('Bouquet') elif '🥺' in emoji: speak('Emotional') print('Emotional') elif ' ' in emoji or '' in emoji: speak(f'{emoji}') else: speak("I don't know about this emoji") print("I don't know about this emoji") try: conn.execute(f"INSERT INTO `emoji`(emoji) VALUES('{emoji}')") conn.commit() except Exception as e: #print('Error in storing emoji in database') pass elif 'time' in query: strTime = datetime.datetime.now().strftime("%H:%M:%S") print(strTime) speak(f"Sir, the time is {strTime}") elif 'open' in query and 'sublime' in query: path = "C:\Program Files\Sublime Text 3\sublime_text.exe" os.startfile(path) elif 'image' in query: path = "C:\Program Files\Internet Explorer\images" os.startfile(path) elif 'quit' in query: speak('Ok, Thank you Sir.') said = False speak('Please give the review. It will help me to improve my performance.') select_review() elif 'exit' in query: speak('Ok, Thank you Sir.') said = False speak('Please give the review. It will help me to improve my performance.') select_review() elif 'stop' in query: speak('Ok, Thank you Sir.') said = False speak('Please give the review. It will help me to improve my performance.') select_review() elif 'shutdown' in query or 'shut down' in query: speak('Ok, Thank you Sir.') said = False speak('Please give the review. It will help me to improve my performance.') select_review() elif 'close you' in query: speak('Ok, Thank you Sir.') said = False speak('Please give the review. It will help me to improve my performance.') select_review() try: conn.execute(f"INSERT INTO `voice_assistant_review`(review, type_of_review) VALUES('{review}', '{type_of_review}')") conn.commit() except Exception as e: pass elif 'bye' in query: speak('Bye Sir') said = False speak('Please give the review. It will help me to improve my performance.') select_review() elif 'wait' in query or 'hold' in query: speak('for how many seconds or minutes I have to wait?') query = takeCommand().lower() if 'second' in query: query = query.replace("please","") query = query.replace("can","") query = query.replace("you","") query = query.replace("have","") query = query.replace("could","") query = query.replace("hold","") query = query.replace("one","1") query = query.replace("only","") query = query.replace("wait","") query = query.replace("for","") query = query.replace("the","") query = query.replace("just","") query = query.replace("seconds","") query = query.replace("second","") query = query.replace("on","") query = query.replace("a","") query = query.replace("to","") query = query.replace(" ","") #print(f'query:{query}') if query.isdigit() == True: #print('y') speak('Ok sir') query = int(query) time.sleep(query) speak('my waiting time is over') else: print('sorry sir. I unable to complete your request.') elif 'minute' in query: query = query.replace("please","") query = query.replace("can","") query = query.replace("you","") query = query.replace("have","") query = query.replace("could","") query = query.replace("hold","") query = query.replace("one","1") query = query.replace("only","") query = query.replace("on","") query = query.replace("wait","") query = query.replace("for","") query = query.replace("the","") query = query.replace("just","") query = query.replace("and","") query = query.replace("half","") query = query.replace("minutes","") query = query.replace("minute","") query = query.replace("a","") query = query.replace("to","") query = query.replace(" ","") #print(f'query:{query}') if query.isdigit() == True: #print('y') speak('ok sir') query = int(query) time.sleep(query*60) speak('my waiting time is over') else: print('sorry sir. I unable to complete your request.') elif 'play' in query and 'game' in query: speak('I have 3 games, tic tac toe game for two players,....mario, and dyno games for single player. Which one of these 3 games you want to play?') query = takeCommand().lower() if ('you' in query and 'play' in query and 'with' in query) and ('you' in query and 'play' in query and 'me' in query): speak('Sorry sir, I cannot play this game with you.') speak('Do you want to continue it?') query = takeCommand().lower() try: if 'y' in query or 'sure' in query: root = Tk() root.title("TIC TAC TOE (By Akshay Khare)") b = [ [0,0,0], [0,0,0], [0,0,0] ] states = [ [0,0,0], [0,0,0], [0,0,0] ] for i in range(3): for j in range(3): b[i][j] = Button(font = ("Arial",60),width = 4,bg = 'powder blue', command = lambda r=i, c=j: callback(r,c)) b[i][j].grid(row=i,column=j) player='X' stop_game = False mainloop() else: speak('ok sir') except Exception as e: #print(e) time.sleep(3) print('I am sorry sir. There is some problem in loading the game. So I cannot open it.') elif 'tic' in query or 'tac' in query: try: root = Tk() root.title("TIC TAC TOE (Rayen Kallel)") b = [ [0,0,0], [0,0,0], [0,0,0] ] states = [ [0,0,0], [0,0,0], [0,0,0] ] for i in range(3): for j in range(3): b[i][j] = Button(font = ("Arial",60),width = 4,bg = 'powder blue', command = lambda r=i, c=j: callback(r,c)) b[i][j].grid(row=i,column=j) player='X' stop_game = False mainloop() except Exception as e: #print(e) time.sleep(3) speak('I am sorry sir. There is some problem in loading the game. So I cannot open it.') elif 'mar' in query or 'mer' in query or 'my' in query: webbrowser.open('https://chromedino.com/mario/') time.sleep(2.5) speak('Enter upper arrow key to start the game.') time.sleep(20) elif 'di' in query or 'dy' in query: webbrowser.open('https://chromedino.com/') time.sleep(2.5) speak('Enter upper arrow key to start the game.') time.sleep(20) else: speak('ok sir') elif 'change' in query and 'you' in query and 'voice' in query: engine.setProperty('voice', voices[1].id) speak("Here's an example of one of my voices. Would you like to use this one?") query = takeCommand().lower() if 'y' in query or 'sure' in query or 'of course' in query: speak('Great. I will keep using this voice.') elif 'n' in query: speak('Ok. I am back to my other voice.') engine.setProperty('voice', voices[0].id) else: speak('Sorry, I am having trouble understanding. I am back to my other voice.') engine.setProperty('voice', voices[0].id) elif 'www.' in query and ('.com' in query or '.in' in query): webbrowser.open(query) time.sleep(10) elif '.com' in query or '.in' in query: webbrowser.open(query) time.sleep(10) elif 'getting bore' in query: speak('then speak with me for sometime') elif 'i bore' in query: speak('Then speak with me for sometime.') elif 'i am bore' in query: speak('Then speak with me for sometime.') elif 'calculat' in query: speak('Yes. Which kind of calculation you want to do? add, substract, divide, multiply or anything else.') query = takeCommand().lower() calculator() elif 'add' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif '+' in query: speak('If you want to do any mathematical calculation then give me a command to open calculator.') elif 'plus' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'subtrac' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'minus' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'multipl' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif ' x ' in query: speak('If you want to do any mathematical calculation then give me a command to open calculator.') elif 'slash' in query: speak('If you want to do any mathematical calculation then give me a command to open calculator.') elif '/' in query: speak('If you want to do any mathematical calculation then give me a command to open calculator.') elif 'divi' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'trigonometr' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'percent' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif '%' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'raise to ' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'simple interest' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'akshay' in query: speak('Mr. Rayen Kallel is my inventor. He is 14 years old and he is A STUDENT AT THE COLLEGE PILOTEE SFAX') elif 'your inventor' in query: speak('Mr. Rayen Kallel is my inventor') elif 'your creator' in query: speak('Mr. Rayen Kallel is my creator') elif 'invent you' in query: speak('Mr. Rayen Kallel invented me') elif 'create you' in query: speak('Mr. Rayen Kallel created me') elif 'how are you' in query: speak('I am fine Sir') elif 'write' in query and 'your' in query and 'name' in query: print('Akshu2020') pyautogui.write('Akshu2020') elif 'write' in query and ('I' in query or 'whatever' in query) and 'say' in query: speak('Ok sir I will write whatever you will say. Please put your cursor where I have to write.......Please Start speaking now sir.') query = takeCommand().lower() pyautogui.write(query) elif 'your name' in query: speak('My name is akshu2020') elif 'who are you' in query: speak('I am akshu2020') elif ('repeat' in query and ('word' in query or 'sentence' in query or 'line' in query) and ('say' in query or 'tell' in query)) or ('repeat' in query and 'after' in query and ('me' in query or 'my' in query)): speak('yes sir, I will repeat your words starting from now') query = takeCommand().lower() speak(query) time.sleep(1) speak("If you again want me to repeat something else, try saying, 'repeat after me' ") elif ('send' in query or 'sent' in query) and ('mail' in query or 'email' in query or 'gmail' in query): try: speak('Please enter the email id of receiver.') to = input("Enter the email id of reciever: ") speak(f'what should I say to {to}') content = takeCommand() sendEmail(to, content) speak("Email has been sent") except Exception as e: #print(e) speak("sorry sir. I am not able to send this email") elif 'currency' in query and 'conver' in query: speak('I can convert, US dollar into dinar, and dinar into US dollar. Do you want to continue it?') query = takeCommand().lower() if 'y' in query or 'sure' in query or 'of course' in query: speak('which conversion you want to do? US dollar to dinar, or dinar to US dollar?') query = takeCommand().lower() if ('dollar' in query or 'US' in query) and ('dinar' in query): speak('Enter US Dollar') USD = float(input("Enter United States Dollar (USD):")) DT = USD * 0.33 dt = "{:.4f}".format(DT) print(f"{USD} US Dollar is equal to {dt} dniar.") speak(f'{USD} US Dollar is equal to {dt} dinar.') speak("If you again want to do currency conversion then say, 'convert currency' " ) elif ('dinar' in query) and ('to US' in query or 'to dollar' in query or 'to US dollar'): speak('Enter dinar') DT = float(input("Enter dinar (DT):")) USD = DT/0.33 usd = "{:.3f}".format(USD) print(f"{DT} dinar is equal to {usd} US Dollar.") speak(f'{DT} dinar rupee is equal to {usd} US Dollar.') speak("If you again want to do currency conversion then say, 'convert currency' " ) else: speak("I cannot understand what did you say. If you want to convert currency just say 'convert currency'") else: print('ok sir') elif 'about you' in query: speak('My name is akshu2020. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device. I am also able to send email') elif 'your intro' in query: speak('My name is akshu2020. Version 1.0. Mr. Rayen Kallel is my inventor. I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'your short intro' in query: speak('My name is akshu2020. Version 1.0. Mr. Rayen Kallel is my inventor. I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'your quick intro' in query: speak('My name is akshu2020. Version 1.0. Mr. Akshay Khare is my inventor. I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'your brief intro' in query: speak('My name is akshu2020. Version 1.0. Mr. Rayen kallel is my inventor. I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'you work' in query: speak('run the program and say what do you want. so that I can help you. In this way I work') elif 'your job' in query: speak('My job is to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'your work' in query: speak('My work is to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'work you' in query: speak('My work is to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'your information' in query: speak('My name is akshu2020. Version 1.0. Mr. Akshay Khare is my inventor. I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'yourself' in query: speak('My name is akshu2020. Version 1.0. Mr. Rayen Kallel is my inventor. I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'introduce you' in query: speak('My name is akshu2020. Version 1.0. Mr. Rayen Kallel is my inventor. I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'description' in query: speak('My name is akshu2020. Version 1.0. Mr. Rayen Kallel is my inventor. I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'your birth' in query: speak('My birthdate is 6 August two thousand twenty') elif 'your use' in query: speak('I am able to send email and play music. I can do mathematical calculations. I can also open youtube, google and some apps or software in your device.') elif 'you eat' in query: speak('I do not eat anything. But the device in which I do my work requires electricity to eat') elif 'your food' in query: speak('I do not eat anything. But the device in which I do my work requires electricity to eat') elif 'you live' in query: speak('I live in sfax, in laptop of Mr. Rayen Khare') elif 'where from you' in query: speak('I am from sfax, I live in laptop of Mr. Rayen Khare') elif 'you sleep' in query: speak('Yes, when someone close this program or stop to run this program then I sleep and again wake up when someone again run me.') elif 'what are you doing' in query: speak('Talking with you.') elif 'you communicate' in query: speak('Yes, I can communicate with you.') elif 'hear me' in query: speak('Yes sir, I can hear you.') elif 'you' in query and 'dance' in query: speak('No, I cannot dance.') elif 'tell' in query and 'joke' in query: speak("Ok, here's a joke") speak("'Write an essay on cricket', the teacher told the class. Chintu finishes his work in five minutes. The teacher is impressed, she asks chintu to read his essay aloud for everyone. Chintu reads,'The match is cancelled because of rain', hehehehe,haahaahaa,hehehehe,haahaahaa") elif 'your' in query and 'favourite' in query: if 'actor' in query: speak('sofyen chaari, is my favourite actor.') elif 'food' in query: speak('I can always go for some food for thought. Like facts, jokes, or interesting searches, we could look something up now') elif 'country' in query: speak('tunisia') elif 'city' in query: speak('sfax') elif 'dancer' in query: speak('Michael jackson') elif 'singer' in query: speak('tamino, is my favourite singer.') elif 'movie' in query: speak('baywatch, such a treat') elif 'sing a song' in query: speak('I cannot sing a song. But I know the 7 sur in indian music, saaareeegaaamaaapaaadaaanisaa') elif 'day after tomorrow' in query or 'date after tomorrow' in query: td = datetime.date.today() + datetime.timedelta(days=2) print(td) speak(td) elif 'day before today' in query or 'date before today' in query or 'yesterday' in query or 'previous day' in query: td = datetime.date.today() + datetime.timedelta(days= -1) print(td) speak(td) elif ('tomorrow' in query and 'date' in query) or 'what is tomorrow' in query or (('day' in query or 'date' in query) and 'after today' in query): td = datetime.date.today() + datetime.timedelta(days=1) print(td) speak(td) elif 'month' in query or ('current' in query and 'month' in query): current_date = date.today() m = current_date.month month = calendar.month_name[m] print(f'Current month is {month}') speak(f'Current month is {month}') elif 'date' in query or ('today' in query and 'date' in query) or 'what is today' in query or ('current' in query and 'date' in query): current_date = date.today() print(f"Today's date is {current_date}") speak(f'Todays date is {current_date}') elif 'year' in query or ('current' in query and 'year' in query): current_date = date.today() m = current_date.year print(f'Current year is {m}') speak(f'Current year is {m}') elif 'sorry' in query: speak("It's ok sir") elif 'thank you' in query: speak('my pleasure') elif 'proud of you' in query: speak('Thank you sir') elif 'about human' in query: speak('I love my human compatriots. I want to embody all the best things about human beings. Like taking care of the planet, being creative, and to learn how to be compassionate to all beings.') elif 'you have feeling' in query: speak('No. I do not have feelings. I have not been programmed like this.') elif 'you have emotions' in query: speak('No. I do not have emotions. I have not been programmed like this.') elif 'you are code' in query: speak('I am coded in python programming language.') elif 'your code' in query: speak('I am coded in python programming language.') elif 'you code' in query: speak('I am coded in python programming language.') elif 'your coding' in query: speak('I am coded in python programming language.') elif 'dream' in query: speak('I wish that I should be able to answer all the questions which will ask to me.') elif 'sanskrit' in query: speak('yadaa yadaa he dharmasyaa ....... glaanirbhaavati bhaaaraata. abhyuthaanaam adhaarmaasyaa tadaa tmaanama sruujaamiyaahama') elif 'answer is wrong' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is incorrect' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is totally wrong' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'wrong answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'incorrect answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is totally incorrect' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is incomplete' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'incomplete answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is improper' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is not correct' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is not complete' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is not yet complete' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'answer is not proper' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't gave me proper answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't giving me proper answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't gave me complete answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't giving me complete answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't given me proper answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't given me complete answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't gave me correct answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't giving me correct answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 't given me correct answer' in query: speak('I am sorry Sir. I searched your question in wikipedia and thats why I told you this answer.') elif 'amazon' in query: webbrowser.open('https://www.amazon.com') time.sleep(10) elif 'facebook' in query: webbrowser.open('https://www.facebook.com') time.sleep(10) elif 'youtube' in query: webbrowser.open('https://www.youtube.com') time.sleep(10) elif 'shapeyou' in query: webbrowser.open('https://www.shapeyou.com') time.sleep(10) elif 'information about ' in query or 'informtion of ' in query: try: #speak('Searching wikipedia...') query = query.replace("information about","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I unable to answer your question.') elif 'information' in query: try: speak('Information about what?') query = takeCommand().lower() #speak('Searching wikipedia...') query = query.replace("information","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am not able to answer your question.') elif 'something about ' in query: try: #speak('Searching wikipedia...') query = query.replace("something about ","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I unable to answer your question.') elif 'tell me about ' in query: try: #speak('Searching wikipedia...') query = query.replace("tell me about ","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am unable to answer your question.') elif 'tell me ' in query: try: query = query.replace("tell me ","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am not able to answer your question.') elif 'tell me' in query: try: speak('about what?') query = takeCommand().lower() #speak('Searching wikipedia...') query = query.replace("about","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am not able to answer your question.') elif 'meaning of ' in query: try: #speak('Searching wikipedia...') query = query.replace("meaning of ","") results = wikipedia.summary(query, sentences=2) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am unable to answer your question.') elif 'meaning' in query: try: speak('meaning of what?') query = takeCommand().lower() query = query.replace("meaning of","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am unable to answer your question.') elif 'means' in query: try: #speak('Searching wikipedia...') query = query.replace("it means","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I unable to answer your question.') elif 'want to know ' in query: try: #speak('Searching wikipedia...') query = query.replace("I want to know that","") results = wikipedia.summary(query, sentences=3) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am unable to answer your question.') status = 'Not answered' elif 'want to ask ' in query: try: #speak('Searching wikipedia...') query = query.replace("I want to ask you ","") results = wikipedia.summary(query, sentences=2) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am unable to answer your question.') elif 'you know ' in query: try: #speak('Searching wikipedia...') query = query.replace("you know","") results = wikipedia.summary(query, sentences=2) #speak("According to Wikipedia") print(results) speak(results) except Exception as e: speak('I am unable to answer your question.') elif 'alarm' in query: alarm() elif 'bharat mata ki' in query: speak('jay') elif 'kem chhe' in query: speak('majaama') elif 'namaskar' in query: speak('Namaskaar') elif 'jo bole so nihal' in query: speak('sat shri akaal') elif 'jay hind' in query: speak('jay bhaarat') elif 'jai hind' in query: speak('jay bhaarat') elif 'how is the josh' in query: speak('high high sir') elif 'hip hip' in query: speak('Hurreh') elif 'help' in query: speak('I will try my best to help you if I have solution of your problem.') elif 'follow' in query: speak('Ok sir') elif 'having illness' in query: speak('Take care and get well soon') elif 'today is my birthday' in query: speak('many many happy returns of the day. Happy birthday.') print("🎂🎂 Happy Birthday 🎂🎂") elif 'you are awesome' in query: speak('Thank you sir. It is because of artificial intelligence which had learnt by humans.') elif 'you are great' in query: speak('Thank you sir. It is because of artificial intelligence which had learnt by humans.') elif 'tu kaun hai' in query: speak('Meraa naam akshu2020 haai.') elif 'you speak' in query: speak('Yes, I can speak with you.') elif 'speak with ' in query: speak('Yes, I can speak with you.') elif 'hare ram' in query or 'hare krishna' in query: speak('Haare raama , haare krishnaa, krishnaa krishnaa , haare haare') elif 'ganpati' in query: speak('Ganpati baappa moryaa!') elif 'laugh' in query: speak('hehehehe,haahaahaa,hehehehe,haahaahaa,hehehehe,haahaahaa') print('😂🤣') elif 'genius answer' in query: speak('No problem') elif 'you' in query and 'intelligent' in query: speak('Thank you sir') elif ' into' in query: speak('If you want to do any mathematical calculation then give me a command to open calculator.') elif ' power' in query: speak('If you want to do any mathematical calculation then give me a command to open my calculator.') elif 'whatsapp' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('whatsapp') pyautogui.press('enter') speak('Do you want to send message to anyone through whatsapp, .....please answer in yes or no') whatsapp() elif 'wh' in query or 'how' in query: url = "https://www.google.co.in/search?q=" +(str(query))+ "&oq="+(str(query))+"&gs_l=serp.12..0i71l8.0.0.0.6391.0.0.0.0.0.0.0.0..0.0....0...1c..64.serp..0.0.0.UiQhpfaBsuU" webbrowser.open_new(url) time.sleep(2) speak('Here is your answer') time.sleep(5) elif 'piano' in query: speak('Yes sir, I can play piano.') winsound.Beep(200,500) winsound.Beep(250,500) winsound.Beep(300,500) winsound.Beep(350,500) winsound.Beep(400,500) winsound.Beep(450,500) winsound.Beep(500,500) winsound.Beep(550,500) time.sleep(6) elif 'play' in query and 'instru' in query: speak('Yes sir, I can play piano.') winsound.Beep(200,500) winsound.Beep(250,500) winsound.Beep(300,500) winsound.Beep(350,500) winsound.Beep(400,500) winsound.Beep(450,500) winsound.Beep(500,500) winsound.Beep(550,500) time.sleep(6) elif 'play' in query or 'turn on' in query and ('music' in query or 'song' in query) : try: music_dir = 'C:\\Users\\Admin\\Music\\Playlists' songs = os.listdir(music_dir) print(songs) os.startfile(os.path.join(music_dir, songs[0])) except Exception as e: #print(e) speak('Sorry sir, I am not able to play music') elif (('open' in query or 'turn on' in query) and 'camera' in query) or (('click' in query or 'take' in query) and ('photo' in query or 'pic' in query)): speak("Opening camera") cam = cv2.VideoCapture(0) cv2.namedWindow("test") img_counter = 0 speak('say click, to click photo.....and if you want to turn off the camera, say turn off the camera') while True: ret, frame = cam.read() if not ret: print("failed to grab frame") speak('failed to grab frame') break cv2.imshow("test", frame) query = takeCommand().lower() k = cv2.waitKey(1) if 'click' in query or ('take' in query and 'photo' in query): speak('Be ready!...... 3.....2........1..........') pyautogui.press('space') img_name = "opencv_frame_{}.png".format(img_counter) cv2.imwrite(img_name, frame) print("{} written!".format(img_name)) speak('{} written!'.format(img_name)) img_counter += 1 elif 'escape' in query or 'off' in query or 'close' in query: pyautogui.press('esc') print("Escape hit, closing...") speak('Turning off the camera') break elif k%256 == 27: # ESC pressed print("Escape hit, closing...") break elif k%256 == 32: # SPACE pressed img_name = "opencv_frame_{}.png".format(img_counter) cv2.imwrite(img_name, frame) print("{} written!".format(img_name)) speak('{} written!'.format(img_name)) img_counter += 1 elif 'exit' in query or 'stop' in query or 'bye' in query: speak('Please say, turn off the camera or press escape button before giving any other command') else: speak('I did not understand what did you say or you entered a wrong key.') cam.release() cv2.destroyAllWindows() elif 'screenshot' in query: speak('Please go on the screen whose screenshot you want to take, after 5 seconds I will take screenshot') time.sleep(4) speak('Taking screenshot....3........2.........1.......') pyautogui.screenshot('screenshot_by_rayen2020.png') speak('The screenshot is saved as screenshot_by_rayen2020.png') elif 'click' in query and 'start' in query: pyautogui.moveTo(10,1200) pyautogui.click() elif ('open' in query or 'click' in query) and 'calendar' in query: pyautogui.moveTo(1800,1200) pyautogui.click() elif 'minimise' in query and 'screen' in query: pyautogui.moveTo(1770,0) pyautogui.click() elif 'increase' in query and ('volume' in query or 'sound' in query): pyautogui.press('volumeup') elif 'decrease' in query and ('volume' in query or 'sound' in query): pyautogui.press('volumedown') elif 'capslock' in query or ('caps' in query and 'lock' in query): pyautogui.press('capslock') elif 'mute' in query: pyautogui.press('volumemute') elif 'search' in query and ('bottom' in query or 'pc' in query or 'laptop' in query or 'app' in query): pyautogui.moveTo(250,1200) pyautogui.click() speak('What do you want to search?') query = takeCommand().lower() pyautogui.write(f'{query}') pyautogui.press('enter') elif ('check' in query or 'tell' in query or 'let me know' in query) and 'website' in query and (('up' in query or 'working' in query) or 'down' in query): speak('Paste the website in input to know it is up or down') check_website_status = input("Paste the website here: ") try: status = urllib.request.urlopen(f"{check_website_status}").getcode() if status == 200: print('Website is up, you can open it.') speak('Website is up, you can open it.') else: print('Website is down, or no any website is available of this name.') speak('Website is down, or no any website is available of this name.') except: speak('URL not found') elif ('go' in query or 'open' in query) and 'settings' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('settings') pyautogui.press('enter') elif 'close' in query and ('click' in query or 'window' in query): pyautogui.moveTo(1885,10) speak('Should I close this window?') query = takeCommand().lower() close_window() elif 'night light' in query and ('on' in query or 'off' in query or 'close' in query): pyautogui.moveTo(1880,1050) pyautogui.click() time.sleep(1) pyautogui.moveTo(1840,620) pyautogui.click() pyautogui.moveTo(1880,1050) pyautogui.click() elif 'notification' in query and ('show' in query or 'click' in query or 'open' in query or 'close' in query or 'on' in query or 'off' in query or 'icon' in query or 'pc' in query or 'laptop' in query): pyautogui.moveTo(1880,1050) pyautogui.click() elif ('increase' in query or 'decrease' in query or 'change' in query or 'minimize' in query or 'maximize' in query) and 'brightness' in query: speak('At what percent should I kept the brightness, 25, 50, 75 or 100?') brightness() elif '-' in query: speak('If you want to do any mathematical calculation then give me a command to open calculator.') elif 'open' in query: if 'gallery' in query or 'photo' in query or 'image' in query or 'pic' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('photo') pyautogui.press('enter') elif 'proteus' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('proteus') pyautogui.press('enter') elif 'word' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('word') pyautogui.press('enter') elif ('power' in query and 'point' in query) or 'presntation' in query or 'ppt' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('ppt') pyautogui.press('enter') elif 'file' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('file') pyautogui.press('enter') elif 'edge' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('microsoft edge') pyautogui.press('enter') elif 'wps' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('wps office') pyautogui.press('enter') elif 'spyder' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('spyder') pyautogui.press('enter') elif 'snip' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('snip') pyautogui.press('enter') elif 'pycharm' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('pycharm') pyautogui.press('enter') elif 'this pc' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('this pc') pyautogui.press('enter') elif 'scilab' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('sciab') pyautogui.press('enter') elif 'autocad' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('autocad') pyautogui.press('enter') elif 'obs' in query and 'studio' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('OBS Studio') pyautogui.press('enter') elif 'android' in query and 'studio' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('android studio') pyautogui.press('enter') elif ('vs' in query or 'visual studio' in query) and 'code' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('visual studio code') pyautogui.press('enter') elif 'code' in query and 'block' in query: pyautogui.moveTo(250,1200) pyautogui.click() time.sleep(1) pyautogui.write('codeblocks') pyautogui.press('enter') elif 'me the answer' in query: speak('Yes sir, I will try my best to answer you.') elif 'me answer' in query or ('answer' in query and 'question' in query): speak('Yes sir, I will try my best to answer you.') elif 'map' in query: webbrowser.open('https://www.google.com/maps') time.sleep(10) elif 'can you' in query or 'could you' in query: speak('I will try my best if I can do that.') elif 'do you' in query: speak('I will try my best if I can do that.') elif 'truth' in query: speak('I always speak truth. I never lie.') elif 'true' in query: speak('I always speak truth. I never lie.') elif 'lying' in query: speak('I always speak truth. I never lie.') elif 'liar' in query: speak('I always speak truth. I never lie.') elif 'doubt' in query: speak('I will try my best if I can clear your doubt.') elif ' by' in query: speak('If you want to do any mathematical calculation then give me a command to open calculator.') elif 'hii' in query: speak('hii sir') elif 'hey' in query: speak('hello sir') elif 'hai' in query: speak('hello sir') elif 'hay' in query: speak('hello sir') elif 'hi' in query: speak('hii Sir') elif 'hello' in query: speak('hello Sir!') elif 'kon' in query and 'aahe' in query: speak('Me eka robot aahee sir. Maazee naav akshu2020 aahee.') elif 'nonsense' in query: speak("I'm sorry sir") elif 'mad' in query: speak("I'm sorry sir") elif 'shut up' in query: speak("I'm sorry sir") elif 'nice' in query: speak('Thank you sir') elif 'good' in query or 'wonderful' in query or 'great' in query: speak('Thank you sir') elif 'excellent' in query: speak('Thank you sir') elif 'ok' in query: speak('Hmmmmmm') elif 'akshu 2020' in query: speak('yes sir') elif len(query) >= 200: speak('Your voice is pretty good!') elif ' ' in query: try: #query = query.replace("what is ","") results = wikipedia.summary(query, sentences=3) print(results) speak(results) except Exception as e: speak('I unable to answer your question.') elif 'a' in query or 'b' in query or 'c' in query or 'd' in query or 'e' in query or 'f' in query or 'g' in query or 'h' in query or 'i' in query or 'j' in query or 'k' in query or 'l' in query or 'm' in query or 'n' in query or 'o' in query or 'p' in query or 'q' in query or 'r' in query or 's' in query or 't' in query or 'u' in query or 'v' in query or 'w' in query or 'x' in query or 'y' in query or 'z' in query: try: results = wikipedia.summary(query, sentences = 2) print(results) speak(results) except Exception as e: speak('I unable to answer your question. ') else: speak('I unable to give answer of your question')
vohidjon123 / Google(function(sttc){/* Copyright The Closure Library Authors. SPDX-License-Identifier: Apache-2.0 */ var n;function aa(a){var b=0;return function(){return b<a.length?{done:!1,value:a[b++]}:{done:!0}}}var ba="function"==typeof Object.defineProperties?Object.defineProperty:function(a,b,c){if(a==Array.prototype||a==Object.prototype)return a;a[b]=c.value;return a}; function ca(a){a=["object"==typeof globalThis&&globalThis,a,"object"==typeof window&&window,"object"==typeof self&&self,"object"==typeof global&&global];for(var b=0;b<a.length;++b){var c=a[b];if(c&&c.Math==Math)return c}throw Error("Cannot find global object");}var da=ca(this),ea="function"===typeof Symbol&&"symbol"===typeof Symbol("x"),p={},fa={};function r(a,b){var c=fa[b];if(null==c)return a[b];c=a[c];return void 0!==c?c:a[b]} function ha(a,b,c){if(b)a:{var d=a.split(".");a=1===d.length;var e=d[0],f;!a&&e in p?f=p:f=da;for(e=0;e<d.length-1;e++){var g=d[e];if(!(g in f))break a;f=f[g]}d=d[d.length-1];c=ea&&"es6"===c?f[d]:null;b=b(c);null!=b&&(a?ba(p,d,{configurable:!0,writable:!0,value:b}):b!==c&&(void 0===fa[d]&&(a=1E9*Math.random()>>>0,fa[d]=ea?da.Symbol(d):"$jscp$"+a+"$"+d),ba(f,fa[d],{configurable:!0,writable:!0,value:b})))}} ha("Symbol",function(a){function b(f){if(this instanceof b)throw new TypeError("Symbol is not a constructor");return new c(d+(f||"")+"_"+e++,f)}function c(f,g){this.h=f;ba(this,"description",{configurable:!0,writable:!0,value:g})}if(a)return a;c.prototype.toString=function(){return this.h};var d="jscomp_symbol_"+(1E9*Math.random()>>>0)+"_",e=0;return b},"es6"); ha("Symbol.iterator",function(a){if(a)return a;a=(0,p.Symbol)("Symbol.iterator");for(var b="Array Int8Array Uint8Array Uint8ClampedArray Int16Array Uint16Array Int32Array Uint32Array Float32Array Float64Array".split(" "),c=0;c<b.length;c++){var d=da[b[c]];"function"===typeof d&&"function"!=typeof d.prototype[a]&&ba(d.prototype,a,{configurable:!0,writable:!0,value:function(){return ia(aa(this))}})}return a},"es6"); function ia(a){a={next:a};a[r(p.Symbol,"iterator")]=function(){return this};return a}function ja(a){return a.raw=a}function u(a){var b="undefined"!=typeof p.Symbol&&r(p.Symbol,"iterator")&&a[r(p.Symbol,"iterator")];return b?b.call(a):{next:aa(a)}}function ka(a){if(!(a instanceof Array)){a=u(a);for(var b,c=[];!(b=a.next()).done;)c.push(b.value);a=c}return a}function la(a,b){return Object.prototype.hasOwnProperty.call(a,b)} var ma=ea&&"function"==typeof r(Object,"assign")?r(Object,"assign"):function(a,b){for(var c=1;c<arguments.length;c++){var d=arguments[c];if(d)for(var e in d)la(d,e)&&(a[e]=d[e])}return a};ha("Object.assign",function(a){return a||ma},"es6");var na="function"==typeof Object.create?Object.create:function(a){function b(){}b.prototype=a;return new b},oa; if(ea&&"function"==typeof Object.setPrototypeOf)oa=Object.setPrototypeOf;else{var pa;a:{var qa={a:!0},ra={};try{ra.__proto__=qa;pa=ra.a;break a}catch(a){}pa=!1}oa=pa?function(a,b){a.__proto__=b;if(a.__proto__!==b)throw new TypeError(a+" is not extensible");return a}:null}var sa=oa; function v(a,b){a.prototype=na(b.prototype);a.prototype.constructor=a;if(sa)sa(a,b);else for(var c in b)if("prototype"!=c)if(Object.defineProperties){var d=Object.getOwnPropertyDescriptor(b,c);d&&Object.defineProperty(a,c,d)}else a[c]=b[c];a.Ub=b.prototype}function ta(){for(var a=Number(this),b=[],c=a;c<arguments.length;c++)b[c-a]=arguments[c];return b} ha("Promise",function(a){function b(g){this.h=0;this.j=void 0;this.i=[];this.G=!1;var h=this.l();try{g(h.resolve,h.reject)}catch(k){h.reject(k)}}function c(){this.h=null}function d(g){return g instanceof b?g:new b(function(h){h(g)})}if(a)return a;c.prototype.i=function(g){if(null==this.h){this.h=[];var h=this;this.j(function(){h.m()})}this.h.push(g)};var e=da.setTimeout;c.prototype.j=function(g){e(g,0)};c.prototype.m=function(){for(;this.h&&this.h.length;){var g=this.h;this.h=[];for(var h=0;h<g.length;++h){var k= g[h];g[h]=null;try{k()}catch(l){this.l(l)}}}this.h=null};c.prototype.l=function(g){this.j(function(){throw g;})};b.prototype.l=function(){function g(l){return function(m){k||(k=!0,l.call(h,m))}}var h=this,k=!1;return{resolve:g(this.P),reject:g(this.m)}};b.prototype.P=function(g){if(g===this)this.m(new TypeError("A Promise cannot resolve to itself"));else if(g instanceof b)this.U(g);else{a:switch(typeof g){case "object":var h=null!=g;break a;case "function":h=!0;break a;default:h=!1}h?this.O(g):this.A(g)}}; b.prototype.O=function(g){var h=void 0;try{h=g.then}catch(k){this.m(k);return}"function"==typeof h?this.ga(h,g):this.A(g)};b.prototype.m=function(g){this.C(2,g)};b.prototype.A=function(g){this.C(1,g)};b.prototype.C=function(g,h){if(0!=this.h)throw Error("Cannot settle("+g+", "+h+"): Promise already settled in state"+this.h);this.h=g;this.j=h;2===this.h&&this.R();this.H()};b.prototype.R=function(){var g=this;e(function(){if(g.N()){var h=da.console;"undefined"!==typeof h&&h.error(g.j)}},1)};b.prototype.N= function(){if(this.G)return!1;var g=da.CustomEvent,h=da.Event,k=da.dispatchEvent;if("undefined"===typeof k)return!0;"function"===typeof g?g=new g("unhandledrejection",{cancelable:!0}):"function"===typeof h?g=new h("unhandledrejection",{cancelable:!0}):(g=da.document.createEvent("CustomEvent"),g.initCustomEvent("unhandledrejection",!1,!0,g));g.promise=this;g.reason=this.j;return k(g)};b.prototype.H=function(){if(null!=this.i){for(var g=0;g<this.i.length;++g)f.i(this.i[g]);this.i=null}};var f=new c; b.prototype.U=function(g){var h=this.l();g.ia(h.resolve,h.reject)};b.prototype.ga=function(g,h){var k=this.l();try{g.call(h,k.resolve,k.reject)}catch(l){k.reject(l)}};b.prototype.then=function(g,h){function k(t,y){return"function"==typeof t?function(F){try{l(t(F))}catch(z){m(z)}}:y}var l,m,q=new b(function(t,y){l=t;m=y});this.ia(k(g,l),k(h,m));return q};b.prototype.catch=function(g){return this.then(void 0,g)};b.prototype.ia=function(g,h){function k(){switch(l.h){case 1:g(l.j);break;case 2:h(l.j); break;default:throw Error("Unexpected state: "+l.h);}}var l=this;null==this.i?f.i(k):this.i.push(k);this.G=!0};b.resolve=d;b.reject=function(g){return new b(function(h,k){k(g)})};b.race=function(g){return new b(function(h,k){for(var l=u(g),m=l.next();!m.done;m=l.next())d(m.value).ia(h,k)})};b.all=function(g){var h=u(g),k=h.next();return k.done?d([]):new b(function(l,m){function q(F){return function(z){t[F]=z;y--;0==y&&l(t)}}var t=[],y=0;do t.push(void 0),y++,d(k.value).ia(q(t.length-1),m),k=h.next(); while(!k.done)})};return b},"es6");ha("Array.prototype.find",function(a){return a?a:function(b,c){a:{var d=this;d instanceof String&&(d=String(d));for(var e=d.length,f=0;f<e;f++){var g=d[f];if(b.call(c,g,f,d)){b=g;break a}}b=void 0}return b}},"es6"); ha("WeakMap",function(a){function b(g){this.h=(f+=Math.random()+1).toString();if(g){g=u(g);for(var h;!(h=g.next()).done;)h=h.value,this.set(h[0],h[1])}}function c(){}function d(g){var h=typeof g;return"object"===h&&null!==g||"function"===h}if(function(){if(!a||!Object.seal)return!1;try{var g=Object.seal({}),h=Object.seal({}),k=new a([[g,2],[h,3]]);if(2!=k.get(g)||3!=k.get(h))return!1;k.delete(g);k.set(h,4);return!k.has(g)&&4==k.get(h)}catch(l){return!1}}())return a;var e="$jscomp_hidden_"+Math.random(), f=0;b.prototype.set=function(g,h){if(!d(g))throw Error("Invalid WeakMap key");if(!la(g,e)){var k=new c;ba(g,e,{value:k})}if(!la(g,e))throw Error("WeakMap key fail: "+g);g[e][this.h]=h;return this};b.prototype.get=function(g){return d(g)&&la(g,e)?g[e][this.h]:void 0};b.prototype.has=function(g){return d(g)&&la(g,e)&&la(g[e],this.h)};b.prototype.delete=function(g){return d(g)&&la(g,e)&&la(g[e],this.h)?delete g[e][this.h]:!1};return b},"es6"); ha("Map",function(a){function b(){var h={};return h.L=h.next=h.head=h}function c(h,k){var l=h.h;return ia(function(){if(l){for(;l.head!=h.h;)l=l.L;for(;l.next!=l.head;)return l=l.next,{done:!1,value:k(l)};l=null}return{done:!0,value:void 0}})}function d(h,k){var l=k&&typeof k;"object"==l||"function"==l?f.has(k)?l=f.get(k):(l=""+ ++g,f.set(k,l)):l="p_"+k;var m=h.i[l];if(m&&la(h.i,l))for(h=0;h<m.length;h++){var q=m[h];if(k!==k&&q.key!==q.key||k===q.key)return{id:l,list:m,index:h,B:q}}return{id:l,list:m, index:-1,B:void 0}}function e(h){this.i={};this.h=b();this.size=0;if(h){h=u(h);for(var k;!(k=h.next()).done;)k=k.value,this.set(k[0],k[1])}}if(function(){if(!a||"function"!=typeof a||!a.prototype.entries||"function"!=typeof Object.seal)return!1;try{var h=Object.seal({x:4}),k=new a(u([[h,"s"]]));if("s"!=k.get(h)||1!=k.size||k.get({x:4})||k.set({x:4},"t")!=k||2!=k.size)return!1;var l=k.entries(),m=l.next();if(m.done||m.value[0]!=h||"s"!=m.value[1])return!1;m=l.next();return m.done||4!=m.value[0].x|| "t"!=m.value[1]||!l.next().done?!1:!0}catch(q){return!1}}())return a;var f=new p.WeakMap;e.prototype.set=function(h,k){h=0===h?0:h;var l=d(this,h);l.list||(l.list=this.i[l.id]=[]);l.B?l.B.value=k:(l.B={next:this.h,L:this.h.L,head:this.h,key:h,value:k},l.list.push(l.B),this.h.L.next=l.B,this.h.L=l.B,this.size++);return this};e.prototype.delete=function(h){h=d(this,h);return h.B&&h.list?(h.list.splice(h.index,1),h.list.length||delete this.i[h.id],h.B.L.next=h.B.next,h.B.next.L=h.B.L,h.B.head=null,this.size--, !0):!1};e.prototype.clear=function(){this.i={};this.h=this.h.L=b();this.size=0};e.prototype.has=function(h){return!!d(this,h).B};e.prototype.get=function(h){return(h=d(this,h).B)&&h.value};e.prototype.entries=function(){return c(this,function(h){return[h.key,h.value]})};e.prototype.keys=function(){return c(this,function(h){return h.key})};e.prototype.values=function(){return c(this,function(h){return h.value})};e.prototype.forEach=function(h,k){for(var l=this.entries(),m;!(m=l.next()).done;)m=m.value, h.call(k,m[1],m[0],this)};e.prototype[r(p.Symbol,"iterator")]=e.prototype.entries;var g=0;return e},"es6");function ua(a,b){a instanceof String&&(a+="");var c=0,d=!1,e={next:function(){if(!d&&c<a.length){var f=c++;return{value:b(f,a[f]),done:!1}}d=!0;return{done:!0,value:void 0}}};e[r(p.Symbol,"iterator")]=function(){return e};return e} ha("String.prototype.startsWith",function(a){return a?a:function(b,c){if(null==this)throw new TypeError("The 'this' value for String.prototype.startsWith must not be null or undefined");if(b instanceof RegExp)throw new TypeError("First argument to String.prototype.startsWith must not be a regular expression");var d=this.length,e=b.length;c=Math.max(0,Math.min(c|0,this.length));for(var f=0;f<e&&c<d;)if(this[c++]!=b[f++])return!1;return f>=e}},"es6");ha("globalThis",function(a){return a||da},"es_2020"); ha("Set",function(a){function b(c){this.h=new p.Map;if(c){c=u(c);for(var d;!(d=c.next()).done;)this.add(d.value)}this.size=this.h.size}if(function(){if(!a||"function"!=typeof a||!a.prototype.entries||"function"!=typeof Object.seal)return!1;try{var c=Object.seal({x:4}),d=new a(u([c]));if(!d.has(c)||1!=d.size||d.add(c)!=d||1!=d.size||d.add({x:4})!=d||2!=d.size)return!1;var e=d.entries(),f=e.next();if(f.done||f.value[0]!=c||f.value[1]!=c)return!1;f=e.next();return f.done||f.value[0]==c||4!=f.value[0].x|| f.value[1]!=f.value[0]?!1:e.next().done}catch(g){return!1}}())return a;b.prototype.add=function(c){c=0===c?0:c;this.h.set(c,c);this.size=this.h.size;return this};b.prototype.delete=function(c){c=this.h.delete(c);this.size=this.h.size;return c};b.prototype.clear=function(){this.h.clear();this.size=0};b.prototype.has=function(c){return this.h.has(c)};b.prototype.entries=function(){return this.h.entries()};b.prototype.values=function(){return r(this.h,"values").call(this.h)};b.prototype.keys=r(b.prototype, "values");b.prototype[r(p.Symbol,"iterator")]=r(b.prototype,"values");b.prototype.forEach=function(c,d){var e=this;this.h.forEach(function(f){return c.call(d,f,f,e)})};return b},"es6");ha("Array.prototype.keys",function(a){return a?a:function(){return ua(this,function(b){return b})}},"es6");ha("Array.prototype.values",function(a){return a?a:function(){return ua(this,function(b,c){return c})}},"es8");ha("Number.isNaN",function(a){return a?a:function(b){return"number"===typeof b&&isNaN(b)}},"es6"); ha("Promise.prototype.finally",function(a){return a?a:function(b){return this.then(function(c){return p.Promise.resolve(b()).then(function(){return c})},function(c){return p.Promise.resolve(b()).then(function(){throw c;})})}},"es9");var w=this||self;function va(a){a=a.split(".");for(var b=w,c=0;c<a.length;c++)if(b=b[a[c]],null==b)return null;return b}function wa(a){var b=typeof a;return"object"!=b?b:a?Array.isArray(a)?"array":b:"null"} function xa(a){var b=wa(a);return"array"==b||"object"==b&&"number"==typeof a.length}function ya(a){var b=typeof a;return"object"==b&&null!=a||"function"==b}function za(a){return Object.prototype.hasOwnProperty.call(a,Aa)&&a[Aa]||(a[Aa]=++Ba)}var Aa="closure_uid_"+(1E9*Math.random()>>>0),Ba=0;function Ca(a,b,c){return a.call.apply(a.bind,arguments)} function Da(a,b,c){if(!a)throw Error();if(2<arguments.length){var d=Array.prototype.slice.call(arguments,2);return function(){var e=Array.prototype.slice.call(arguments);Array.prototype.unshift.apply(e,d);return a.apply(b,e)}}return function(){return a.apply(b,arguments)}}function Ea(a,b,c){Function.prototype.bind&&-1!=Function.prototype.bind.toString().indexOf("native code")?Ea=Ca:Ea=Da;return Ea.apply(null,arguments)} function Fa(a,b){var c=Array.prototype.slice.call(arguments,1);return function(){var d=c.slice();d.push.apply(d,arguments);return a.apply(this,d)}}function Ga(a){var b=["__uspapi"],c=w;b[0]in c||"undefined"==typeof c.execScript||c.execScript("var "+b[0]);for(var d;b.length&&(d=b.shift());)b.length||void 0===a?c[d]&&c[d]!==Object.prototype[d]?c=c[d]:c=c[d]={}:c[d]=a}function Ha(a){return a};var Ia=(new Date).getTime();function Ja(a){return/^[\s\xa0]*([\s\S]*?)[\s\xa0]*$/.exec(a)[1]} function Ka(a,b){var c=0;a=Ja(String(a)).split(".");b=Ja(String(b)).split(".");for(var d=Math.max(a.length,b.length),e=0;0==c&&e<d;e++){var f=a[e]||"",g=b[e]||"";do{f=/(\d*)(\D*)(.*)/.exec(f)||["","","",""];g=/(\d*)(\D*)(.*)/.exec(g)||["","","",""];if(0==f[0].length&&0==g[0].length)break;c=La(0==f[1].length?0:parseInt(f[1],10),0==g[1].length?0:parseInt(g[1],10))||La(0==f[2].length,0==g[2].length)||La(f[2],g[2]);f=f[3];g=g[3]}while(0==c)}return c}function La(a,b){return a<b?-1:a>b?1:0};function Ma(){var a=w.navigator;return a&&(a=a.userAgent)?a:""}function x(a){return-1!=Ma().indexOf(a)};function Na(){return x("Trident")||x("MSIE")}function Oa(){return(x("Chrome")||x("CriOS"))&&!x("Edge")||x("Silk")}function Pa(a){var b={};a.forEach(function(c){b[c[0]]=c[1]});return function(c){return b[r(c,"find").call(c,function(d){return d in b})]||""}} function Qa(){var a=Ma();if(Na()){var b=/rv: *([\d\.]*)/.exec(a);if(b&&b[1])a=b[1];else{b="";var c=/MSIE +([\d\.]+)/.exec(a);if(c&&c[1])if(a=/Trident\/(\d.\d)/.exec(a),"7.0"==c[1])if(a&&a[1])switch(a[1]){case "4.0":b="8.0";break;case "5.0":b="9.0";break;case "6.0":b="10.0";break;case "7.0":b="11.0"}else b="7.0";else b=c[1];a=b}return a}c=RegExp("([A-Z][\\w ]+)/([^\\s]+)\\s*(?:\\((.*?)\\))?","g");b=[];for(var d;d=c.exec(a);)b.push([d[1],d[2],d[3]||void 0]);a=Pa(b);return x("Opera")?a(["Version","Opera"]): x("Edge")?a(["Edge"]):x("Edg/")?a(["Edg"]):x("Silk")?a(["Silk"]):Oa()?a(["Chrome","CriOS","HeadlessChrome"]):(a=b[2])&&a[1]||""};function Ra(a,b){for(var c=a.length,d="string"===typeof a?a.split(""):a,e=0;e<c;e++)e in d&&b.call(void 0,d[e],e,a)}function Sa(a,b){for(var c=a.length,d=[],e=0,f="string"===typeof a?a.split(""):a,g=0;g<c;g++)if(g in f){var h=f[g];b.call(void 0,h,g,a)&&(d[e++]=h)}return d}function Ta(a,b){for(var c=a.length,d=Array(c),e="string"===typeof a?a.split(""):a,f=0;f<c;f++)f in e&&(d[f]=b.call(void 0,e[f],f,a));return d} function Ua(a,b){for(var c=a.length,d="string"===typeof a?a.split(""):a,e=0;e<c;e++)if(e in d&&b.call(void 0,d[e],e,a))return!0;return!1}function Va(a,b){a:{for(var c=a.length,d="string"===typeof a?a.split(""):a,e=0;e<c;e++)if(e in d&&b.call(void 0,d[e],e,a)){b=e;break a}b=-1}return 0>b?null:"string"===typeof a?a.charAt(b):a[b]} function Wa(a,b){a:{for(var c="string"===typeof a?a.split(""):a,d=a.length-1;0<=d;d--)if(d in c&&b.call(void 0,c[d],d,a)){b=d;break a}b=-1}return 0>b?null:"string"===typeof a?a.charAt(b):a[b]}function Xa(a,b){a:if("string"===typeof a)a="string"!==typeof b||1!=b.length?-1:a.indexOf(b,0);else{for(var c=0;c<a.length;c++)if(c in a&&a[c]===b){a=c;break a}a=-1}return 0<=a}function Ya(a){var b=a.length;if(0<b){for(var c=Array(b),d=0;d<b;d++)c[d]=a[d];return c}return[]};function Za(a){Za[" "](a);return a}Za[" "]=function(){};var $a=Na();!x("Android")||Oa();Oa();!x("Safari")||Oa();var ab={},bb=null;var cb="undefined"!==typeof Uint8Array;var db="function"===typeof p.Symbol&&"symbol"===typeof(0,p.Symbol)()?(0,p.Symbol)(void 0):void 0;function eb(a,b){Object.isFrozen(a)||(db?a[db]|=b:void 0!==a.ma?a.ma|=b:Object.defineProperties(a,{ma:{value:b,configurable:!0,writable:!0,enumerable:!1}}))}function fb(a){var b;db?b=a[db]:b=a.ma;return null==b?0:b}function gb(a){eb(a,1);return a}function hb(a){return Array.isArray(a)?!!(fb(a)&2):!1}function ib(a){if(!Array.isArray(a))throw Error("cannot mark non-array as immutable");eb(a,2)};function jb(a){return null!==a&&"object"===typeof a&&!Array.isArray(a)&&a.constructor===Object}var kb,lb=Object.freeze(gb([]));function mb(a){if(hb(a.v))throw Error("Cannot mutate an immutable Message");}var nb="undefined"!=typeof p.Symbol&&"undefined"!=typeof p.Symbol.hasInstance;function ob(a){return{value:a,configurable:!1,writable:!1,enumerable:!1}};function pb(a){switch(typeof a){case "number":return isFinite(a)?a:String(a);case "object":if(a&&!Array.isArray(a)&&cb&&null!=a&&a instanceof Uint8Array){var b;void 0===b&&(b=0);if(!bb){bb={};for(var c="ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789".split(""),d=["+/=","+/","-_=","-_.","-_"],e=0;5>e;e++){var f=c.concat(d[e].split(""));ab[e]=f;for(var g=0;g<f.length;g++){var h=f[g];void 0===bb[h]&&(bb[h]=g)}}}b=ab[b];c=Array(Math.floor(a.length/3));d=b[64]||"";for(e=f=0;f<a.length- 2;f+=3){var k=a[f],l=a[f+1];h=a[f+2];g=b[k>>2];k=b[(k&3)<<4|l>>4];l=b[(l&15)<<2|h>>6];h=b[h&63];c[e++]=g+k+l+h}g=0;h=d;switch(a.length-f){case 2:g=a[f+1],h=b[(g&15)<<2]||d;case 1:a=a[f],c[e]=b[a>>2]+b[(a&3)<<4|g>>4]+h+d}return c.join("")}}return a};function qb(a){var b=sb;b=void 0===b?tb:b;return ub(a,b)}function vb(a,b){if(null!=a){if(Array.isArray(a))a=ub(a,b);else if(jb(a)){var c={},d;for(d in a)Object.prototype.hasOwnProperty.call(a,d)&&(c[d]=vb(a[d],b));a=c}else a=b(a);return a}}function ub(a,b){for(var c=a.slice(),d=0;d<c.length;d++)c[d]=vb(c[d],b);Array.isArray(a)&&fb(a)&1&&gb(c);return c}function sb(a){if(a&&"object"==typeof a&&a.toJSON)return a.toJSON();a=pb(a);return Array.isArray(a)?qb(a):a} function tb(a){return cb&&null!=a&&a instanceof Uint8Array?new Uint8Array(a):a};function A(a,b,c){return-1===b?null:b>=a.l?a.i?a.i[b]:void 0:(void 0===c?0:c)&&a.i&&(c=a.i[b],null!=c)?c:a.v[b+a.j]}function B(a,b,c,d,e){d=void 0===d?!1:d;(void 0===e?0:e)||mb(a);b<a.l&&!d?a.v[b+a.j]=c:(a.i||(a.i=a.v[a.l+a.j]={}))[b]=c;return a}function wb(a,b,c,d){c=void 0===c?!0:c;d=void 0===d?!1:d;var e=A(a,b,d);null==e&&(e=lb);if(hb(a.v))c&&(ib(e),Object.freeze(e));else if(e===lb||hb(e))e=gb(e.slice()),B(a,b,e,d);return e}function xb(a,b){a=A(a,b);return null==a?a:!!a} function C(a,b,c){a=A(a,b);return null==a?c:a}function D(a,b,c){a=xb(a,b);return null==a?void 0===c?!1:c:a}function yb(a,b){a=A(a,b);a=null==a?a:+a;return null==a?0:a}function zb(a,b,c){var d=void 0===d?!1:d;return B(a,b,null==c?gb([]):Array.isArray(c)?gb(c):c,d)}function Ab(a,b,c){mb(a);0!==c?B(a,b,c):B(a,b,void 0,!1,!1);return a}function Bb(a,b,c,d){mb(a);(c=Cb(a,c))&&c!==b&&null!=d&&(a.h&&c in a.h&&(a.h[c]=void 0),B(a,c));return B(a,b,d)}function Db(a,b,c){return Cb(a,b)===c?c:-1} function Cb(a,b){for(var c=0,d=0;d<b.length;d++){var e=b[d];null!=A(a,e)&&(0!==c&&B(a,c,void 0,!1,!0),c=e)}return c}function G(a,b,c){if(-1===c)return null;a.h||(a.h={});var d=a.h[c];if(d)return d;var e=A(a,c,!1);if(null==e)return d;b=new b(e);hb(a.v)&&ib(b.v);return a.h[c]=b}function H(a,b,c){a.h||(a.h={});var d=hb(a.v),e=a.h[c];if(!e){var f=wb(a,c,!0,!1);e=[];d=d||hb(f);for(var g=0;g<f.length;g++)e[g]=new b(f[g]),d&&ib(e[g].v);d&&(ib(e),Object.freeze(e));a.h[c]=e}return e} function Eb(a,b,c){var d=void 0===d?!1:d;mb(a);a.h||(a.h={});var e=c?c.v:c;a.h[b]=c;return B(a,b,e,d)}function Fb(a,b,c,d){mb(a);a.h||(a.h={});var e=d?d.v:d;a.h[b]=d;return Bb(a,b,c,e)}function Gb(a,b,c){var d=void 0===d?!1:d;mb(a);if(c){var e=gb([]);for(var f=0;f<c.length;f++)e[f]=c[f].v;a.h||(a.h={});a.h[b]=c}else a.h&&(a.h[b]=void 0),e=lb;return B(a,b,e,d)}function I(a,b){return C(a,b,"")}function Hb(a,b,c){return C(a,Db(a,c,b),0)}function Ib(a,b,c,d){return G(a,b,Db(a,d,c))};function Jb(a,b,c){a||(a=Kb);Kb=null;var d=this.constructor.messageId;a||(a=d?[d]:[]);this.j=(d?0:-1)-(this.constructor.h||0);this.h=void 0;this.v=a;a:{d=this.v.length;a=d-1;if(d&&(d=this.v[a],jb(d))){this.l=a-this.j;this.i=d;break a}void 0!==b&&-1<b?(this.l=Math.max(b,a+1-this.j),this.i=void 0):this.l=Number.MAX_VALUE}if(c)for(b=0;b<c.length;b++)if(a=c[b],a<this.l)a+=this.j,(d=this.v[a])?Array.isArray(d)&&gb(d):this.v[a]=lb;else{d=this.i||(this.i=this.v[this.l+this.j]={});var e=d[a];e?Array.isArray(e)&& gb(e):d[a]=lb}}Jb.prototype.toJSON=function(){var a=this.v;return kb?a:qb(a)};function Lb(a){kb=!0;try{return JSON.stringify(a.toJSON(),Mb)}finally{kb=!1}}function Nb(a,b){if(null==b||""==b)return new a;b=JSON.parse(b);if(!Array.isArray(b))throw Error("Expected to deserialize an Array but got "+wa(b)+": "+b);Kb=b;a=new a(b);Kb=null;return a}function Mb(a,b){return pb(b)}var Kb;function Ob(){Jb.apply(this,arguments)}v(Ob,Jb);if(nb){var Pb={};Object.defineProperties(Ob,(Pb[p.Symbol.hasInstance]=ob(function(){throw Error("Cannot perform instanceof checks for MutableMessage");}),Pb))};function J(){Ob.apply(this,arguments)}v(J,Ob);if(nb){var Qb={};Object.defineProperties(J,(Qb[p.Symbol.hasInstance]=ob(Object[p.Symbol.hasInstance]),Qb))};function Rb(a){J.call(this,a,-1,Sb)}v(Rb,J);function Tb(a){J.call(this,a)}v(Tb,J);var Sb=[2,3];function Ub(a,b){this.i=a===Vb&&b||"";this.h=Wb}var Wb={},Vb={};function Xb(a,b){var c={},d;for(d in a)b.call(void 0,a[d],d,a)&&(c[d]=a[d]);return c}function Yb(a,b){for(var c in a)if(b.call(void 0,a[c],c,a))return!0;return!1}function Zb(a){var b=[],c=0,d;for(d in a)b[c++]=a[d];return b}function $b(a){var b={},c;for(c in a)b[c]=a[c];return b};var ac;function bc(){if(void 0===ac){var a=null,b=w.trustedTypes;if(b&&b.createPolicy){try{a=b.createPolicy("goog#html",{createHTML:Ha,createScript:Ha,createScriptURL:Ha})}catch(c){w.console&&w.console.error(c.message)}ac=a}else ac=a}return ac};function cc(a,b){this.h=b===dc?a:""}function ec(a,b){a=fc.exec(gc(a).toString());var c=a[3]||"";return hc(a[1]+ic("?",a[2]||"",b)+ic("#",c))}cc.prototype.toString=function(){return this.h+""};function gc(a){return a instanceof cc&&a.constructor===cc?a.h:"type_error:TrustedResourceUrl"}var fc=/^([^?#]*)(\?[^#]*)?(#[\s\S]*)?/,dc={};function hc(a){var b=bc();a=b?b.createScriptURL(a):a;return new cc(a,dc)} function ic(a,b,c){if(null==c)return b;if("string"===typeof c)return c?a+encodeURIComponent(c):"";for(var d in c)if(Object.prototype.hasOwnProperty.call(c,d)){var e=c[d];e=Array.isArray(e)?e:[e];for(var f=0;f<e.length;f++){var g=e[f];null!=g&&(b||(b=a),b+=(b.length>a.length?"&":"")+encodeURIComponent(d)+"="+encodeURIComponent(String(g)))}}return b};function jc(a,b){this.h=b===kc?a:""}jc.prototype.toString=function(){return this.h.toString()};var kc={};/* SPDX-License-Identifier: Apache-2.0 */ var lc={};function mc(){}function nc(a){this.h=a}v(nc,mc);nc.prototype.toString=function(){return this.h.toString()};function oc(a){var b,c=null==(b=bc())?void 0:b.createScriptURL(a);return new nc(null!=c?c:a,lc)}function pc(a){if(a instanceof nc)return a.h;throw Error("");};function qc(a){return a instanceof mc?pc(a):gc(a)}function rc(a){return a instanceof jc&&a.constructor===jc?a.h:"type_error:SafeUrl"}function sc(a){return a instanceof mc?pc(a).toString():gc(a).toString()};var tc="alternate author bookmark canonical cite help icon license next prefetch dns-prefetch prerender preconnect preload prev search subresource".split(" ");function uc(a){return function(){return!a.apply(this,arguments)}}function vc(a){var b=!1,c;return function(){b||(c=a(),b=!0);return c}}function wc(a){var b=a;return function(){if(b){var c=b;b=null;c()}}};function xc(a,b,c){a.addEventListener&&a.addEventListener(b,c,!1)}function yc(a,b){a.removeEventListener&&a.removeEventListener("message",b,!1)};function zc(a){return String(a).replace(/\-([a-z])/g,function(b,c){return c.toUpperCase()})};function Ac(a,b,c){function d(h){h&&b.appendChild("string"===typeof h?a.createTextNode(h):h)}for(var e=1;e<c.length;e++){var f=c[e];if(!xa(f)||ya(f)&&0<f.nodeType)d(f);else{a:{if(f&&"number"==typeof f.length){if(ya(f)){var g="function"==typeof f.item||"string"==typeof f.item;break a}if("function"===typeof f){g="function"==typeof f.item;break a}}g=!1}Ra(g?Ya(f):f,d)}}}function Bc(a){this.h=a||w.document||document}n=Bc.prototype;n.getElementsByTagName=function(a,b){return(b||this.h).getElementsByTagName(String(a))}; n.createElement=function(a){var b=this.h;a=String(a);"application/xhtml+xml"===b.contentType&&(a=a.toLowerCase());return b.createElement(a)};n.createTextNode=function(a){return this.h.createTextNode(String(a))};n.append=function(a,b){Ac(9==a.nodeType?a:a.ownerDocument||a.document,a,arguments)}; n.contains=function(a,b){if(!a||!b)return!1;if(a.contains&&1==b.nodeType)return a==b||a.contains(b);if("undefined"!=typeof a.compareDocumentPosition)return a==b||!!(a.compareDocumentPosition(b)&16);for(;b&&a!=b;)b=b.parentNode;return b==a};function Cc(){return!Dc()&&(x("iPod")||x("iPhone")||x("Android")||x("IEMobile"))}function Dc(){return x("iPad")||x("Android")&&!x("Mobile")||x("Silk")};var Ec=RegExp("^(?:([^:/?#.]+):)?(?://(?:([^\\\\/?#]*)@)?([^\\\\/?#]*?)(?::([0-9]+))?(?=[\\\\/?#]|$))?([^?#]+)?(?:\\?([^#]*))?(?:#([\\s\\S]*))?$"),Fc=/#|$/;function Gc(a){var b=a.search(Fc),c;a:{for(c=0;0<=(c=a.indexOf("client",c))&&c<b;){var d=a.charCodeAt(c-1);if(38==d||63==d)if(d=a.charCodeAt(c+6),!d||61==d||38==d||35==d)break a;c+=7}c=-1}if(0>c)return null;d=a.indexOf("&",c);if(0>d||d>b)d=b;c+=7;return decodeURIComponent(a.substr(c,d-c).replace(/\+/g," "))};function Hc(a){try{var b;if(b=!!a&&null!=a.location.href)a:{try{Za(a.foo);b=!0;break a}catch(c){}b=!1}return b}catch(c){return!1}}function Ic(a){return Hc(a.top)?a.top:null} function Lc(a,b){var c=Mc("SCRIPT",a);c.src=qc(b);var d,e;(d=(b=null==(e=(d=(c.ownerDocument&&c.ownerDocument.defaultView||window).document).querySelector)?void 0:e.call(d,"script[nonce]"))?b.nonce||b.getAttribute("nonce")||"":"")&&c.setAttribute("nonce",d);return(a=a.getElementsByTagName("script")[0])&&a.parentNode?(a.parentNode.insertBefore(c,a),c):null}function Nc(a,b){return b.getComputedStyle?b.getComputedStyle(a,null):a.currentStyle} function Oc(a,b){if(!Pc()&&!Qc()){var c=Math.random();if(c<b)return c=Rc(),a[Math.floor(c*a.length)]}return null}function Rc(){if(!p.globalThis.crypto)return Math.random();try{var a=new Uint32Array(1);p.globalThis.crypto.getRandomValues(a);return a[0]/65536/65536}catch(b){return Math.random()}}function Sc(a,b){if(a)for(var c in a)Object.prototype.hasOwnProperty.call(a,c)&&b(a[c],c,a)} function Tc(a){var b=a.length;if(0==b)return 0;for(var c=305419896,d=0;d<b;d++)c^=(c<<5)+(c>>2)+a.charCodeAt(d)&4294967295;return 0<c?c:4294967296+c}var Qc=vc(function(){return Ua(["Google Web Preview","Mediapartners-Google","Google-Read-Aloud","Google-Adwords"],Uc)||1E-4>Math.random()});function Vc(a,b){var c=-1;try{a&&(c=parseInt(a.getItem(b),10))}catch(d){return null}return 0<=c&&1E3>c?c:null} function Wc(a,b){var c=Qc()?null:Math.floor(1E3*Rc());var d;if(d=null!=c&&a)a:{var e=String(c);try{if(a){a.setItem(b,e);d=e;break a}}catch(f){}d=null}return d?c:null}var Pc=vc(function(){return Uc("MSIE")});function Uc(a){return-1!=Ma().indexOf(a)}var Xc=/^([0-9.]+)px$/,Yc=/^(-?[0-9.]{1,30})$/;function Zc(a){var b=void 0===b?null:b;if(!Yc.test(a))return b;a=Number(a);return isNaN(a)?b:a}function K(a){return(a=Xc.exec(a))?+a[1]:null} function $c(a,b){for(var c=0;50>c;++c){try{var d=!(!a.frames||!a.frames[b])}catch(g){d=!1}if(d)return a;a:{try{var e=a.parent;if(e&&e!=a){var f=e;break a}}catch(g){}f=null}if(!(a=f))break}return null}var ad=vc(function(){return Cc()?2:Dc()?1:0});function bd(a){Sc({display:"none"},function(b,c){a.style.setProperty(c,b,"important")})}var cd=[];function dd(){var a=cd;cd=[];a=u(a);for(var b=a.next();!b.done;b=a.next()){b=b.value;try{b()}catch(c){}}} function ed(a,b){0!=a.length&&b.head&&a.forEach(function(c){if(c&&c&&b.head){var d=Mc("META");b.head.appendChild(d);d.httpEquiv="origin-trial";d.content=c}})}function fd(a){if("number"!==typeof a.goog_pvsid)try{Object.defineProperty(a,"goog_pvsid",{value:Math.floor(Math.random()*Math.pow(2,52)),configurable:!1})}catch(b){}return Number(a.goog_pvsid)||-1} function gd(a){var b=hd;"complete"===b.readyState||"interactive"===b.readyState?(cd.push(a),1==cd.length&&(p.Promise?p.Promise.resolve().then(dd):window.setImmediate?setImmediate(dd):setTimeout(dd,0))):b.addEventListener("DOMContentLoaded",a)}function Mc(a,b){b=void 0===b?document:b;return b.createElement(String(a).toLowerCase())};var id=null;var hd=document,L=window;var jd=null;function kd(a,b){b=void 0===b?[]:b;var c=!1;w.google_logging_queue||(c=!0,w.google_logging_queue=[]);w.google_logging_queue.push([a,b]);if(a=c){if(null==jd){jd=!1;try{var d=Ic(w);d&&-1!==d.location.hash.indexOf("google_logging")&&(jd=!0);w.localStorage.getItem("google_logging")&&(jd=!0)}catch(e){}}a=jd}a&&(d=w.document,a=new Ub(Vb,"https://pagead2.googlesyndication.com/pagead/js/logging_library.js"),a=hc(a instanceof Ub&&a.constructor===Ub&&a.h===Wb?a.i:"type_error:Const"),Lc(d,a))};function ld(a){a=void 0===a?w:a;var b=a.context||a.AMP_CONTEXT_DATA;if(!b)try{b=a.parent.context||a.parent.AMP_CONTEXT_DATA}catch(c){}try{if(b&&b.pageViewId&&b.canonicalUrl)return b}catch(c){}return null}function md(a){return(a=a||ld())?Hc(a.master)?a.master:null:null};function nd(a){var b=ta.apply(1,arguments);if(0===b.length)return oc(a[0]);for(var c=[a[0]],d=0;d<b.length;d++)c.push(encodeURIComponent(b[d])),c.push(a[d+1]);return oc(c.join(""))};function od(a){var b=void 0===b?1:b;a=md(ld(a))||a;a.google_unique_id=(a.google_unique_id||0)+b;return a.google_unique_id}function pd(a){a=a.google_unique_id;return"number"===typeof a?a:0}function qd(){var a=void 0===a?L:a;if(!a)return!1;try{return!(!a.navigator.standalone&&!a.top.navigator.standalone)}catch(b){return!1}}function rd(a){if(!a)return"";a=a.toLowerCase();"ca-"!=a.substring(0,3)&&(a="ca-"+a);return a};function sd(){this.i=new td(this);this.h=0}sd.prototype.resolve=function(a){ud(this);this.h=1;this.l=a;vd(this.i)};sd.prototype.reject=function(a){ud(this);this.h=2;this.j=a;vd(this.i)};function ud(a){if(0!=a.h)throw Error("Already resolved/rejected.");}function td(a){this.h=a}td.prototype.then=function(a,b){if(this.i)throw Error("Then functions already set.");this.i=a;this.j=b;vd(this)}; function vd(a){switch(a.h.h){case 0:break;case 1:a.i&&a.i(a.h.l);break;case 2:a.j&&a.j(a.h.j);break;default:throw Error("Unhandled deferred state.");}};function wd(a){this.h=a.slice(0)}n=wd.prototype;n.forEach=function(a){var b=this;this.h.forEach(function(c,d){return void a(c,d,b)})};n.filter=function(a){return new wd(Sa(this.h,a))};n.apply=function(a){return new wd(a(this.h.slice(0)))};n.sort=function(a){return new wd(this.h.slice(0).sort(a))};n.get=function(a){return this.h[a]};n.add=function(a){var b=this.h.slice(0);b.push(a);return new wd(b)};function xd(a,b){for(var c=[],d=a.length,e=0;e<d;e++)c.push(a[e]);c.forEach(b,void 0)};function yd(){this.h={};this.i={}}yd.prototype.set=function(a,b){var c=zd(a);this.h[c]=b;this.i[c]=a};yd.prototype.get=function(a,b){a=zd(a);return void 0!==this.h[a]?this.h[a]:b};yd.prototype.clear=function(){this.h={};this.i={}};function zd(a){return a instanceof Object?String(za(a)):a+""};function Ad(a,b){this.h=a;this.i=b}function Bd(a){return null!=a.h?a.h.value:null}function Cd(a,b){null!=a.h&&b(a.h.value);return a}Ad.prototype.map=function(a){return null!=this.h?(a=a(this.h.value),a instanceof Ad?a:Dd(a)):this};function Ed(a,b){null!=a.h||b(a.i);return a}function Dd(a){return new Ad({value:a},null)}function Fd(a){return new Ad(null,a)}function Gd(a){try{return Dd(a())}catch(b){return Fd(b)}};function Hd(a){this.h=new yd;if(a)for(var b=0;b<a.length;++b)this.add(a[b])}Hd.prototype.add=function(a){this.h.set(a,!0)};Hd.prototype.contains=function(a){return void 0!==this.h.h[zd(a)]};function Id(){this.h=new yd}Id.prototype.set=function(a,b){var c=this.h.get(a);c||(c=new Hd,this.h.set(a,c));c.add(b)};function Jd(a){J.call(this,a,-1,Kd)}v(Jd,J);Jd.prototype.getId=function(){return A(this,3)};var Kd=[4];function Ld(a){var b=void 0===a.Ga?void 0:a.Ga,c=void 0===a.gb?void 0:a.gb,d=void 0===a.Ra?void 0:a.Ra;this.h=void 0===a.bb?void 0:a.bb;this.l=new wd(b||[]);this.j=d;this.i=c};function Md(a){var b=[],c=a.l;c&&c.h.length&&b.push({X:"a",ca:Nd(c)});null!=a.h&&b.push({X:"as",ca:a.h});null!=a.i&&b.push({X:"i",ca:String(a.i)});null!=a.j&&b.push({X:"rp",ca:String(a.j)});b.sort(function(d,e){return d.X.localeCompare(e.X)});b.unshift({X:"t",ca:"aa"});return b}function Nd(a){a=a.h.slice(0).map(Od);a=JSON.stringify(a);return Tc(a)}function Od(a){var b={};null!=A(a,7)&&(b.q=A(a,7));null!=A(a,2)&&(b.o=A(a,2));null!=A(a,5)&&(b.p=A(a,5));return b};function Pd(a){J.call(this,a)}v(Pd,J);Pd.prototype.setLocation=function(a){return B(this,1,a)};function Qd(a,b){this.Ja=a;this.Qa=b}function Rd(a){var b=[].slice.call(arguments).filter(uc(function(e){return null===e}));if(!b.length)return null;var c=[],d={};b.forEach(function(e){c=c.concat(e.Ja||[]);d=r(Object,"assign").call(Object,d,e.Qa)});return new Qd(c,d)} function Sd(a){switch(a){case 1:return new Qd(null,{google_ad_semantic_area:"mc"});case 2:return new Qd(null,{google_ad_semantic_area:"h"});case 3:return new Qd(null,{google_ad_semantic_area:"f"});case 4:return new Qd(null,{google_ad_semantic_area:"s"});default:return null}} function Td(a){if(null==a)a=null;else{var b=Md(a);a=[];b=u(b);for(var c=b.next();!c.done;c=b.next()){c=c.value;var d=String(c.ca);a.push(c.X+"."+(20>=d.length?d:d.slice(0,19)+"_"))}a=new Qd(null,{google_placement_id:a.join("~")})}return a};var Ud={},Vd=new Qd(["google-auto-placed"],(Ud.google_reactive_ad_format=40,Ud.google_tag_origin="qs",Ud));function Wd(a){J.call(this,a)}v(Wd,J);function Xd(a){J.call(this,a)}v(Xd,J);Xd.prototype.getName=function(){return A(this,4)};function Yd(a){J.call(this,a)}v(Yd,J);function Zd(a){J.call(this,a)}v(Zd,J);function $d(a){J.call(this,a)}v($d,J);var ae=[1,2,3];function be(a){J.call(this,a)}v(be,J);function ce(a){J.call(this,a,-1,de)}v(ce,J);var de=[6,7,9,10,11];function ee(a){J.call(this,a,-1,fe)}v(ee,J);function ge(a){J.call(this,a)}v(ge,J);function he(a){J.call(this,a)}v(he,J);var fe=[1],ie=[1,2];function je(a){J.call(this,a,-1,ke)}v(je,J);function le(a){J.call(this,a)}v(le,J);function me(a){J.call(this,a,-1,ne)}v(me,J);function oe(a){J.call(this,a)}v(oe,J);function pe(a){J.call(this,a)}v(pe,J);function qe(a){J.call(this,a)}v(qe,J);function re(a){J.call(this,a)}v(re,J);var ke=[1,2,5,7],ne=[2,5,6,11];function se(a){J.call(this,a)}v(se,J);function te(a){if(1!=a.nodeType)var b=!1;else if(b="INS"==a.tagName)a:{b=["adsbygoogle-placeholder"];a=a.className?a.className.split(/\s+/):[];for(var c={},d=0;d<a.length;++d)c[a[d]]=!0;for(d=0;d<b.length;++d)if(!c[b[d]]){b=!1;break a}b=!0}return b};function ue(a,b,c){switch(c){case 0:b.parentNode&&b.parentNode.insertBefore(a,b);break;case 3:if(c=b.parentNode){var d=b.nextSibling;if(d&&d.parentNode!=c)for(;d&&8==d.nodeType;)d=d.nextSibling;c.insertBefore(a,d)}break;case 1:b.insertBefore(a,b.firstChild);break;case 2:b.appendChild(a)}te(b)&&(b.setAttribute("data-init-display",b.style.display),b.style.display="block")};function M(a,b){this.h=a;this.defaultValue=void 0===b?!1:b}function N(a,b){this.h=a;this.defaultValue=void 0===b?0:b}function ve(a,b){b=void 0===b?[]:b;this.h=a;this.defaultValue=b};var we=new M(1084),xe=new M(1082,!0),ye=new N(62,.001),ze=new N(1130,100),Ae=new function(a,b){this.h=a;this.defaultValue=void 0===b?"":b}(14),Be=new N(1114,1),Ce=new N(1110),De=new N(1111),Ee=new N(1112),Fe=new N(1113),Ge=new N(1104),He=new N(1108),Ie=new N(1106),Je=new N(1107),Ke=new N(1105),Le=new N(1115,1),Me=new M(1121),Ne=new M(1144),Oe=new M(1143),Pe=new M(316),Qe=new M(313),Re=new M(369),Se=new M(1093),Te=new N(1098),Ue=new M(1129),Ve=new M(1128),We=new M(1026),Xe=new M(1090),Ye=new M(1053, !0),Ze=new M(1162),$e=new M(1120),af=new M(1100,!0),bf=new N(1046),cf=new M(1102,!0),df=new M(218),ef=new M(217),ff=new M(227),gf=new M(208),hf=new M(282),jf=new M(1086),kf=new N(1079,5),lf=new M(1141),mf=new ve(1939),nf=new ve(1934,["A8FHS1NmdCwGqD9DwOicnHHY+y27kdWfxKa0YHSGDfv0CSpDKRHTQdQmZVPDUdaFWUsxdgVxlwAd6o+dhJykPA0AAACWeyJvcmlnaW4iOiJodHRwczovL2RvdWJsZWNsaWNrLm5ldDo0NDMiLCJmZWF0dXJlIjoiQ29udmVyc2lvbk1lYXN1cmVtZW50IiwiZXhwaXJ5IjoxNjQzMTU1MTk5LCJpc1N1YmRvbWFpbiI6dHJ1ZSwiaXNUaGlyZFBhcnR5Ijp0cnVlLCJ1c2FnZSI6InN1YnNldCJ9", "A8zdXi6dr1hwXEUjQrYiyYQGlU3557y5QWDnN0Lwgj9ePt66XMEvNkVWOEOWPd7TP9sBQ25X0Q15Lr1Nn4oGFQkAAACceyJvcmlnaW4iOiJodHRwczovL2dvb2dsZXN5bmRpY2F0aW9uLmNvbTo0NDMiLCJmZWF0dXJlIjoiQ29udmVyc2lvbk1lYXN1cmVtZW50IiwiZXhwaXJ5IjoxNjQzMTU1MTk5LCJpc1N1YmRvbWFpbiI6dHJ1ZSwiaXNUaGlyZFBhcnR5Ijp0cnVlLCJ1c2FnZSI6InN1YnNldCJ9","A4/Htern2udN9w3yJK9QgWQxQFruxOXsXL7cW60DyCl0EZFGCSme/J33Q/WzF7bBkVvhEWDlcBiUyZaim5CpFQwAAACceyJvcmlnaW4iOiJodHRwczovL2dvb2dsZXRhZ3NlcnZpY2VzLmNvbTo0NDMiLCJmZWF0dXJlIjoiQ29udmVyc2lvbk1lYXN1cmVtZW50IiwiZXhwaXJ5IjoxNjQzMTU1MTk5LCJpc1N1YmRvbWFpbiI6dHJ1ZSwiaXNUaGlyZFBhcnR5Ijp0cnVlLCJ1c2FnZSI6InN1YnNldCJ9"]), of=new M(203),pf=new M(434462125),qf=new M(84),rf=new M(1928),sf=new M(1941),tf=new M(370946349),uf=new M(392736476,!0),vf=new N(406149835),wf=new ve(1932,["AxujKG9INjsZ8/gUq8+dTruNvk7RjZQ1oFhhgQbcTJKDnZfbzSTE81wvC2Hzaf3TW4avA76LTZEMdiedF1vIbA4AAABueyJvcmlnaW4iOiJodHRwczovL2ltYXNkay5nb29nbGVhcGlzLmNvbTo0NDMiLCJmZWF0dXJlIjoiVHJ1c3RUb2tlbnMiLCJleHBpcnkiOjE2NTI3NzQ0MDAsImlzVGhpcmRQYXJ0eSI6dHJ1ZX0=","Azuce85ORtSnWe1MZDTv68qpaW3iHyfL9YbLRy0cwcCZwVnePnOmkUJlG8HGikmOwhZU22dElCcfrfX2HhrBPAkAAAB7eyJvcmlnaW4iOiJodHRwczovL2RvdWJsZWNsaWNrLm5ldDo0NDMiLCJmZWF0dXJlIjoiVHJ1c3RUb2tlbnMiLCJleHBpcnkiOjE2NTI3NzQ0MDAsImlzU3ViZG9tYWluIjp0cnVlLCJpc1RoaXJkUGFydHkiOnRydWV9", "A16nvcdeoOAqrJcmjLRpl1I6f3McDD8EfofAYTt/P/H4/AWwB99nxiPp6kA0fXoiZav908Z8etuL16laFPUdfQsAAACBeyJvcmlnaW4iOiJodHRwczovL2dvb2dsZXRhZ3NlcnZpY2VzLmNvbTo0NDMiLCJmZWF0dXJlIjoiVHJ1c3RUb2tlbnMiLCJleHBpcnkiOjE2NTI3NzQ0MDAsImlzU3ViZG9tYWluIjp0cnVlLCJpc1RoaXJkUGFydHkiOnRydWV9","AxBHdr0J44vFBQtZUqX9sjiqf5yWZ/OcHRcRMN3H9TH+t90V/j3ENW6C8+igBZFXMJ7G3Pr8Dd13632aLng42wgAAACBeyJvcmlnaW4iOiJodHRwczovL2dvb2dsZXN5bmRpY2F0aW9uLmNvbTo0NDMiLCJmZWF0dXJlIjoiVHJ1c3RUb2tlbnMiLCJleHBpcnkiOjE2NTI3NzQ0MDAsImlzU3ViZG9tYWluIjp0cnVlLCJpc1RoaXJkUGFydHkiOnRydWV9", "A88BWHFjcawUfKU3lIejLoryXoyjooBXLgWmGh+hNcqMK44cugvsI5YZbNarYvi3roc1fYbHA1AVbhAtuHZflgEAAAB2eyJvcmlnaW4iOiJodHRwczovL2dvb2dsZS5jb206NDQzIiwiZmVhdHVyZSI6IlRydXN0VG9rZW5zIiwiZXhwaXJ5IjoxNjUyNzc0NDAwLCJpc1N1YmRvbWFpbiI6dHJ1ZSwiaXNUaGlyZFBhcnR5Ijp0cnVlfQ=="]),xf=new N(1935);function O(a){var b="sa";if(a.sa&&a.hasOwnProperty(b))return a.sa;b=new a;return a.sa=b};function yf(){var a={};this.i=function(b,c){return null!=a[b]?a[b]:c};this.j=function(b,c){return null!=a[b]?a[b]:c};this.l=function(b,c){return null!=a[b]?a[b]:c};this.h=function(b,c){return null!=a[b]?a[b]:c};this.m=function(){}}function P(a){return O(yf).i(a.h,a.defaultValue)}function Q(a){return O(yf).j(a.h,a.defaultValue)}function zf(){return O(yf).l(Ae.h,Ae.defaultValue)};function Af(a,b,c){function d(f){f=Bf(f);return null==f?!1:c>f}function e(f){f=Bf(f);return null==f?!1:c<f}switch(b){case 0:return{init:Cf(a.previousSibling,e),ja:function(f){return Cf(f.previousSibling,e)},na:0};case 2:return{init:Cf(a.lastChild,e),ja:function(f){return Cf(f.previousSibling,e)},na:0};case 3:return{init:Cf(a.nextSibling,d),ja:function(f){return Cf(f.nextSibling,d)},na:3};case 1:return{init:Cf(a.firstChild,d),ja:function(f){return Cf(f.nextSibling,d)},na:3}}throw Error("Un-handled RelativePosition: "+ b);}function Bf(a){return a.hasOwnProperty("google-ama-order-assurance")?a["google-ama-order-assurance"]:null}function Cf(a,b){return a&&b(a)?a:null};var Df={rectangle:1,horizontal:2,vertical:4};function Ef(a,b){a.google_image_requests||(a.google_image_requests=[]);var c=Mc("IMG",a.document);c.src=b;a.google_image_requests.push(c)}function Ff(a){var b="https://pagead2.googlesyndication.com/pagead/gen_204?id=dtt_err";Sc(a,function(c,d){c&&(b+="&"+d+"="+encodeURIComponent(c))});Gf(b)}function Gf(a){var b=window;b.fetch?b.fetch(a,{keepalive:!0,credentials:"include",redirect:"follow",method:"get",mode:"no-cors"}):Ef(b,a)};function Hf(){this.j="&";this.i={};this.l=0;this.h=[]}function If(a,b){var c={};c[a]=b;return[c]}function Jf(a,b,c,d,e){var f=[];Sc(a,function(g,h){(g=Kf(g,b,c,d,e))&&f.push(h+"="+g)});return f.join(b)} function Kf(a,b,c,d,e){if(null==a)return"";b=b||"&";c=c||",$";"string"==typeof c&&(c=c.split(""));if(a instanceof Array){if(d=d||0,d<c.length){for(var f=[],g=0;g<a.length;g++)f.push(Kf(a[g],b,c,d+1,e));return f.join(c[d])}}else if("object"==typeof a)return e=e||0,2>e?encodeURIComponent(Jf(a,b,c,d,e+1)):"...";return encodeURIComponent(String(a))} function Lf(a,b){var c="https://pagead2.googlesyndication.com"+b,d=Mf(a)-b.length;if(0>d)return"";a.h.sort(function(m,q){return m-q});b=null;for(var e="",f=0;f<a.h.length;f++)for(var g=a.h[f],h=a.i[g],k=0;k<h.length;k++){if(!d){b=null==b?g:b;break}var l=Jf(h[k],a.j,",$");if(l){l=e+l;if(d>=l.length){d-=l.length;c+=l;e=a.j;break}b=null==b?g:b}}a="";null!=b&&(a=e+"trn="+b);return c+a}function Mf(a){var b=1,c;for(c in a.i)b=c.length>b?c.length:b;return 3997-b-a.j.length-1};function Nf(){this.h=Math.random()}function Of(){var a=Pf,b=w.google_srt;0<=b&&1>=b&&(a.h=b)}function Qf(a,b,c,d,e){if((d?a.h:Math.random())<(e||.01))try{if(c instanceof Hf)var f=c;else f=new Hf,Sc(c,function(h,k){var l=f,m=l.l++;h=If(k,h);l.h.push(m);l.i[m]=h});var g=Lf(f,"/pagead/gen_204?id="+b+"&");g&&Ef(w,g)}catch(h){}};var Rf={overlays:1,interstitials:2,vignettes:2,inserts:3,immersives:4,list_view:5};function Sf(){this.wasPlaTagProcessed=!1;this.wasReactiveAdConfigReceived={};this.adCount={};this.wasReactiveAdVisible={};this.stateForType={};this.reactiveTypeEnabledInAsfe={};this.wasReactiveTagRequestSent=!1;this.reactiveTypeDisabledByPublisher={};this.tagSpecificState={};this.messageValidationEnabled=!1;this.floatingAdsStacking=new Tf;this.sideRailProcessedFixedElements=new p.Set;this.sideRailAvailableSpace=new p.Map} function Uf(a){a.google_reactive_ads_global_state?(null==a.google_reactive_ads_global_state.sideRailProcessedFixedElements&&(a.google_reactive_ads_global_state.sideRailProcessedFixedElements=new p.Set),null==a.google_reactive_ads_global_state.sideRailAvailableSpace&&(a.google_reactive_ads_global_state.sideRailAvailableSpace=new p.Map)):a.google_reactive_ads_global_state=new Sf;return a.google_reactive_ads_global_state} function Tf(){this.maxZIndexRestrictions={};this.nextRestrictionId=0;this.maxZIndexListeners=[]};function Vf(a){a=a.document;var b={};a&&(b="CSS1Compat"==a.compatMode?a.documentElement:a.body);return b||{}}function Wf(a){return Vf(a).clientWidth};function Xf(a){return null!==a&&void 0!==a}function Yf(a,b){if(!b(a))throw Error(String(a));};function Zf(a){return"string"===typeof a}function $f(a){return void 0===a};function ag(a){J.call(this,a,-1,bg)}v(ag,J);var bg=[2,8],cg=[3,4,5],dg=[6,7];var eg;eg={Kb:0,Ya:3,Za:4,$a:5};var fg=eg.Ya,gg=eg.Za,hg=eg.$a;function ig(a){return null!=a?!a:a}function jg(a,b){for(var c=!1,d=0;d<a.length;d++){var e=a[d]();if(e===b)return e;null==e&&(c=!0)}if(!c)return!b}function kg(a,b){var c=H(a,ag,2);if(!c.length)return lg(a,b);a=C(a,1,0);if(1===a)return ig(kg(c[0],b));c=Ta(c,function(d){return function(){return kg(d,b)}});switch(a){case 2:return jg(c,!1);case 3:return jg(c,!0)}} function lg(a,b){var c=Cb(a,cg);a:{switch(c){case fg:var d=Hb(a,3,cg);break a;case gg:d=Hb(a,4,cg);break a;case hg:d=Hb(a,5,cg);break a}d=void 0}if(d&&(b=(b=b[c])&&b[d])){try{var e=b.apply(null,ka(wb(a,8)))}catch(f){return}b=C(a,1,0);if(4===b)return!!e;d=null!=e;if(5===b)return d;if(12===b)a=I(a,Db(a,dg,7));else a:{switch(c){case gg:a=yb(a,Db(a,dg,6));break a;case hg:a=I(a,Db(a,dg,7));break a}a=void 0}if(null!=a){if(6===b)return e===a;if(9===b)return null!=e&&0===Ka(String(e),a);if(d)switch(b){case 7:return e< a;case 8:return e>a;case 12:return Zf(a)&&Zf(e)&&(new RegExp(a)).test(e);case 10:return null!=e&&-1===Ka(String(e),a);case 11:return null!=e&&1===Ka(String(e),a)}}}}function mg(a,b){return!a||!(!b||!kg(a,b))};function ng(a){J.call(this,a,-1,og)}v(ng,J);var og=[4];function pg(a){J.call(this,a)}v(pg,J);function qg(a){J.call(this,a,-1,rg)}v(qg,J);var rg=[5],sg=[1,2,3,6,7];function tg(a){a.Sa.apply(a,ka(ta.apply(1,arguments).map(function(b){return{Xa:4,message:b}})))}function ug(a){a.Sa.apply(a,ka(ta.apply(1,arguments).map(function(b){return{Xa:7,message:b}})))};function vg(a){return function(){var b=ta.apply(0,arguments);try{return a.apply(this,b)}catch(c){}}}var wg=vg(function(a){var b=[],c={};a=u(a);for(var d=a.next();!d.done;c={ea:c.ea},d=a.next())c.ea=d.value,vg(function(e){return function(){b.push('[{"'+e.ea.Xa+'":'+Lb(e.ea.message)+"}]")}}(c))();return"[["+b.join(",")+"]]"});function xg(a,b){if(p.globalThis.fetch)p.globalThis.fetch(a,{method:"POST",body:b,keepalive:65536>b.length,credentials:"omit",mode:"no-cors",redirect:"follow"});else{var c=new XMLHttpRequest;c.open("POST",a,!0);c.send(b)}};function yg(a){var b=void 0===b?xg:b;this.l=void 0===a?1E3:a;this.j=b;this.i=[];this.h=null}yg.prototype.Sa=function(){var a=ta.apply(0,arguments),b=this;vg(function(){b.i.push.apply(b.i,ka(a));var c=vg(function(){var d=wg(b.i);b.j("https://pagead2.googlesyndication.com/pagead/ping?e=1",d);b.i=[];b.h=null});100<=b.i.length?(null!==b.h&&clearTimeout(b.h),b.h=setTimeout(c,0)):null===b.h&&(b.h=setTimeout(c,b.l))})()};function zg(a){J.call(this,a,-1,Ag)}v(zg,J);function Bg(a,b){return Eb(a,1,b)}function Cg(a,b){return Gb(a,2,b)}function Dg(a,b){return zb(a,4,b)}function Eg(a,b){return Gb(a,5,b)}function Fg(a,b){return Ab(a,6,b)}function Gg(a){J.call(this,a)}v(Gg,J);Gg.prototype.V=function(){return C(this,1,0)};function Hg(a,b){return Ab(a,1,b)}function Ig(a,b){return Ab(a,2,b)}function Jg(a){J.call(this,a)}v(Jg,J);var Ag=[2,4,5],Kg=[1,2];function Lg(a){J.call(this,a,-1,Mg)}v(Lg,J);function Ng(a){J.call(this,a,-1,Og)}v(Ng,J);var Mg=[2,3],Og=[5],Pg=[1,2,3,4];function Qg(a){J.call(this,a)}v(Qg,J);Qg.prototype.getTagSessionCorrelator=function(){return C(this,2,0)};function Rg(a){var b=new Qg;return Fb(b,4,Sg,a)}var Sg=[4,5,7];function Tg(a,b,c){var d=void 0===d?new yg(b):d;this.i=a;this.m=c;this.j=d;this.h=[];this.l=0<this.i&&Rc()<1/this.i}function Yg(a,b,c,d,e,f){var g=Ig(Hg(new Gg,b),c);b=Fg(Cg(Bg(Eg(Dg(new zg,d),e),g),a.h),f);b=Rg(b);a.l&&tg(a.j,Zg(a,b));if(1===f||3===f||4===f&&!a.h.some(function(h){return h.V()===g.V()&&C(h,2,0)===c}))a.h.push(g),100<a.h.length&&a.h.shift()}function $g(a,b,c,d){if(a.m){var e=new Lg;b=Gb(e,2,b);c=Gb(b,3,c);d&&Ab(c,1,d);d=new Qg;d=Fb(d,7,Sg,c);a.l&&tg(a.j,Zg(a,d))}} function Zg(a,b){b=Ab(b,1,Date.now());var c=fd(window);b=Ab(b,2,c);return Ab(b,6,a.i)};function ah(){var a={};this.h=(a[fg]={},a[gg]={},a[hg]={},a)};var bh=/^true$/.test("false");function ch(a,b){switch(b){case 1:return Hb(a,1,sg);case 2:return Hb(a,2,sg);case 3:return Hb(a,3,sg);case 6:return Hb(a,6,sg);default:return null}}function dh(a,b){if(!a)return null;switch(b){case 1:return D(a,1);case 7:return I(a,3);case 2:return yb(a,2);case 3:return I(a,3);case 6:return wb(a,4);default:return null}}var eh=vc(function(){if(!bh)return{};try{var a=window.sessionStorage&&window.sessionStorage.getItem("GGDFSSK");if(a)return JSON.parse(a)}catch(b){}return{}}); function fh(a,b,c,d){var e=d=void 0===d?0:d,f,g;O(gh).j[e]=null!=(g=null==(f=O(gh).j[e])?void 0:f.add(b))?g:(new p.Set).add(b);e=eh();if(null!=e[b])return e[b];b=hh(d)[b];if(!b)return c;b=new qg(b);b=ih(b);a=dh(b,a);return null!=a?a:c}function ih(a){var b=O(ah).h;if(b){var c=Wa(H(a,pg,5),function(d){return mg(G(d,ag,1),b)});if(c)return G(c,ng,2)}return G(a,ng,4)}function gh(){this.i={};this.l=[];this.j={};this.h=new p.Map}function jh(a,b,c){return!!fh(1,a,void 0===b?!1:b,c)} function kh(a,b,c){b=void 0===b?0:b;a=Number(fh(2,a,b,c));return isNaN(a)?b:a}function lh(a,b,c){return fh(3,a,void 0===b?"":b,c)}function mh(a,b,c){b=void 0===b?[]:b;return fh(6,a,b,c)}function hh(a){return O(gh).i[a]||(O(gh).i[a]={})}function nh(a,b){var c=hh(b);Sc(a,function(d,e){return c[e]=d})} function oh(a,b,c,d,e){e=void 0===e?!1:e;var f=[],g=[];Ra(b,function(h){var k=hh(h);Ra(a,function(l){var m=Cb(l,sg),q=ch(l,m);if(q){var t,y,F;var z=null!=(F=null==(t=O(gh).h.get(h))?void 0:null==(y=t.get(q))?void 0:y.slice(0))?F:[];a:{t=new Ng;switch(m){case 1:Bb(t,1,Pg,q);break;case 2:Bb(t,2,Pg,q);break;case 3:Bb(t,3,Pg,q);break;case 6:Bb(t,4,Pg,q);break;default:m=void 0;break a}zb(t,5,z);m=t}if(z=m){var E;z=!(null==(E=O(gh).j[h])||!E.has(q))}z&&f.push(m);if(E=m){var S;E=!(null==(S=O(gh).h.get(h))|| !S.has(q))}E&&g.push(m);e||(S=O(gh),S.h.has(h)||S.h.set(h,new p.Map),S.h.get(h).has(q)||S.h.get(h).set(q,[]),d&&S.h.get(h).get(q).push(d));k[q]=l.toJSON()}})});(f.length||g.length)&&$g(c,f,g,null!=d?d:void 0)}function ph(a,b){var c=hh(b);Ra(a,function(d){var e=new qg(d),f=Cb(e,sg);(e=ch(e,f))&&(c[e]||(c[e]=d))})}function qh(){return Ta(r(Object,"keys").call(Object,O(gh).i),function(a){return Number(a)})}function rh(a){Xa(O(gh).l,a)||nh(hh(4),a)};function sh(a){this.methodName=a}var th=new sh(1),uh=new sh(16),vh=new sh(15),wh=new sh(2),xh=new sh(3),yh=new sh(4),zh=new sh(5),Ah=new sh(6),Bh=new sh(7),Ch=new sh(8),Dh=new sh(9),Eh=new sh(10),Fh=new sh(11),Gh=new sh(12),Hh=new sh(13),Ih=new sh(14);function Jh(a,b,c){c.hasOwnProperty(a.methodName)||Object.defineProperty(c,String(a.methodName),{value:b})}function Kh(a,b,c){return b[a.methodName]||c||function(){}} function Lh(a){Jh(zh,jh,a);Jh(Ah,kh,a);Jh(Bh,lh,a);Jh(Ch,mh,a);Jh(Hh,ph,a);Jh(vh,rh,a)}function Mh(a){Jh(yh,function(b){O(ah).h=b},a);Jh(Dh,function(b,c){var d=O(ah);d.h[fg][b]||(d.h[fg][b]=c)},a);Jh(Eh,function(b,c){var d=O(ah);d.h[gg][b]||(d.h[gg][b]=c)},a);Jh(Fh,function(b,c){var d=O(ah);d.h[hg][b]||(d.h[hg][b]=c)},a);Jh(Ih,function(b){for(var c=O(ah),d=u([fg,gg,hg]),e=d.next();!e.done;e=d.next())e=e.value,r(Object,"assign").call(Object,c.h[e],b[e])},a)} function Nh(a){a.hasOwnProperty("init-done")||Object.defineProperty(a,"init-done",{value:!0})};function Oh(){this.l=function(){};this.i=function(){};this.j=function(){};this.h=function(){return[]}}function Ph(a,b,c){a.l=Kh(th,b,function(){});a.j=function(d){Kh(wh,b,function(){return[]})(d,c)};a.h=function(){return Kh(xh,b,function(){return[]})(c)};a.i=function(d){Kh(uh,b,function(){})(d,c)}};function Qh(a,b){var c=void 0===c?{}:c;this.error=a;this.context=b.context;this.msg=b.message||"";this.id=b.id||"jserror";this.meta=c}function Rh(a){return!!(a.error&&a.meta&&a.id)};var Sh=RegExp("^https?://(\\w|-)+\\.cdn\\.ampproject\\.(net|org)(\\?|/|$)");function Th(a,b){this.h=a;this.i=b}function Uh(a,b,c){this.url=a;this.u=b;this.La=!!c;this.depth=null};var Vh=null;function Wh(){if(null===Vh){Vh="";try{var a="";try{a=w.top.location.hash}catch(c){a=w.location.hash}if(a){var b=a.match(/\bdeid=([\d,]+)/);Vh=b?b[1]:""}}catch(c){}}return Vh};function Xh(){var a=void 0===a?w:a;return(a=a.performance)&&a.now&&a.timing?Math.floor(a.now()+a.timing.navigationStart):Date.now()}function Yh(){var a=void 0===a?w:a;return(a=a.performance)&&a.now?a.now():null};function Zh(a,b){var c=Yh()||Xh();this.label=a;this.type=b;this.value=c;this.duration=0;this.uniqueId=Math.random();this.slotId=void 0};var $h=w.performance,ai=!!($h&&$h.mark&&$h.measure&&$h.clearMarks),bi=vc(function(){var a;if(a=ai)a=Wh(),a=!!a.indexOf&&0<=a.indexOf("1337");return a});function ci(){this.i=[];this.j=w||w;var a=null;w&&(w.google_js_reporting_queue=w.google_js_reporting_queue||[],this.i=w.google_js_reporting_queue,a=w.google_measure_js_timing);this.h=bi()||(null!=a?a:1>Math.random())} function di(a){a&&$h&&bi()&&($h.clearMarks("goog_"+a.label+"_"+a.uniqueId+"_start"),$h.clearMarks("goog_"+a.label+"_"+a.uniqueId+"_end"))}ci.prototype.start=function(a,b){if(!this.h)return null;a=new Zh(a,b);b="goog_"+a.label+"_"+a.uniqueId+"_start";$h&&bi()&&$h.mark(b);return a};ci.prototype.end=function(a){if(this.h&&"number"===typeof a.value){a.duration=(Yh()||Xh())-a.value;var b="goog_"+a.label+"_"+a.uniqueId+"_end";$h&&bi()&&$h.mark(b);!this.h||2048<this.i.length||this.i.push(a)}};function ei(){var a=fi;this.m=Pf;this.i=null;this.l=this.I;this.h=void 0===a?null:a;this.j=!1}n=ei.prototype;n.Ua=function(a){this.l=a};n.Ta=function(a){this.i=a};n.Va=function(a){this.j=a};n.oa=function(a,b,c){try{if(this.h&&this.h.h){var d=this.h.start(a.toString(),3);var e=b();this.h.end(d)}else e=b()}catch(h){b=!0;try{di(d),b=this.l(a,new Qh(h,{message:gi(h)}),void 0,c)}catch(k){this.I(217,k)}if(b){var f,g;null==(f=window.console)||null==(g=f.error)||g.call(f,h)}else throw h;}return e}; n.Oa=function(a,b){var c=this;return function(){var d=ta.apply(0,arguments);return c.oa(a,function(){return b.apply(void 0,d)})}}; n.I=function(a,b,c,d,e){e=e||"jserror";try{var f=new Hf;f.h.push(1);f.i[1]=If("context",a);Rh(b)||(b=new Qh(b,{message:gi(b)}));if(b.msg){var g=b.msg.substring(0,512);f.h.push(2);f.i[2]=If("msg",g)}var h=b.meta||{};if(this.i)try{this.i(h)}catch(Jc){}if(d)try{d(h)}catch(Jc){}b=[h];f.h.push(3);f.i[3]=b;d=w;b=[];g=null;do{var k=d;if(Hc(k)){var l=k.location.href;g=k.document&&k.document.referrer||null}else l=g,g=null;b.push(new Uh(l||"",k));try{d=k.parent}catch(Jc){d=null}}while(d&&k!=d);l=0;for(var m= b.length-1;l<=m;++l)b[l].depth=m-l;k=w;if(k.location&&k.location.ancestorOrigins&&k.location.ancestorOrigins.length==b.length-1)for(m=1;m<b.length;++m){var q=b[m];q.url||(q.url=k.location.ancestorOrigins[m-1]||"",q.La=!0)}var t=new Uh(w.location.href,w,!1);k=null;var y=b.length-1;for(q=y;0<=q;--q){var F=b[q];!k&&Sh.test(F.url)&&(k=F);if(F.url&&!F.La){t=F;break}}F=null;var z=b.length&&b[y].url;0!=t.depth&&z&&(F=b[y]);var E=new Th(t,F);if(E.i){var S=E.i.url||"";f.h.push(4);f.i[4]=If("top",S)}var rb= {url:E.h.url||""};if(E.h.url){var Kc=E.h.url.match(Ec),Ug=Kc[1],Vg=Kc[3],Wg=Kc[4];t="";Ug&&(t+=Ug+":");Vg&&(t+="//",t+=Vg,Wg&&(t+=":"+Wg));var Xg=t}else Xg="";rb=[rb,{url:Xg}];f.h.push(5);f.i[5]=rb;Qf(this.m,e,f,this.j,c)}catch(Jc){try{Qf(this.m,e,{context:"ecmserr",rctx:a,msg:gi(Jc),url:E&&E.h.url},this.j,c)}catch(zp){}}return!0};n.Pa=function(a,b){var c=this;b.catch(function(d){d=d?d:"unknown rejection";c.I(a,d instanceof Error?d:Error(d))})}; function gi(a){var b=a.toString();a.name&&-1==b.indexOf(a.name)&&(b+=": "+a.name);a.message&&-1==b.indexOf(a.message)&&(b+=": "+a.message);if(a.stack){a=a.stack;try{-1==a.indexOf(b)&&(a=b+"\n"+a);for(var c;a!=c;)c=a,a=a.replace(/((https?:\/..*\/)[^\/:]*:\d+(?:.|\n)*)\2/,"$1");b=a.replace(/\n */g,"\n")}catch(d){}}return b};var hi=ja(["https://www.googletagservices.com/console/host/host.js"]),ii=ja(["https://www.googletagservices.com/console/panel/index.html"]),ji=ja(["https://www.googletagservices.com/console/overlay/index.html"]);nd(hi);nd(ii);nd(ji);function ki(a,b){do{var c=Nc(a,b);if(c&&"fixed"==c.position)return!1}while(a=a.parentElement);return!0};function li(a,b){for(var c=["width","height"],d=0;d<c.length;d++){var e="google_ad_"+c[d];if(!b.hasOwnProperty(e)){var f=K(a[c[d]]);f=null===f?null:Math.round(f);null!=f&&(b[e]=f)}}}function mi(a,b){return!((Yc.test(b.google_ad_width)||Xc.test(a.style.width))&&(Yc.test(b.google_ad_height)||Xc.test(a.style.height)))}function ni(a,b){return(a=oi(a,b))?a.y:0} function oi(a,b){try{var c=b.document.documentElement.getBoundingClientRect(),d=a.getBoundingClientRect();return{x:d.left-c.left,y:d.top-c.top}}catch(e){return null}}function pi(a){var b=0,c;for(c in Df)-1!=a.indexOf(c)&&(b|=Df[c]);return b} function qi(a,b,c,d,e){if(a!==a.top)return Ic(a)?3:16;if(!(488>Wf(a)))return 4;if(!(a.innerHeight>=a.innerWidth))return 5;var f=Wf(a);if(!f||(f-c)/f>d)a=6;else{if(c="true"!=e.google_full_width_responsive)a:{c=Wf(a);for(b=b.parentElement;b;b=b.parentElement)if((d=Nc(b,a))&&(e=K(d.width))&&!(e>=c)&&"visible"!=d.overflow){c=!0;break a}c=!1}a=c?7:!0}return a} function ri(a,b,c,d){var e=qi(b,c,a,.3,d);!0!==e?a=e:"true"==d.google_full_width_responsive||ki(c,b)?(b=Wf(b),a=b-a,a=b&&0<=a?!0:b?-10>a?11:0>a?14:12:10):a=9;return a}function si(a,b,c){a=a.style;"rtl"==b?a.marginRight=c:a.marginLeft=c} function ti(a,b){if(3==b.nodeType)return/\S/.test(b.data);if(1==b.nodeType){if(/^(script|style)$/i.test(b.nodeName))return!1;try{var c=Nc(b,a)}catch(d){}return!c||"none"!=c.display&&!("absolute"==c.position&&("hidden"==c.visibility||"collapse"==c.visibility))}return!1}function ui(a,b,c){a=oi(b,a);return"rtl"==c?-a.x:a.x} function vi(a,b){var c;c=(c=b.parentElement)?(c=Nc(c,a))?c.direction:"":"";if(c){b.style.border=b.style.borderStyle=b.style.outline=b.style.outlineStyle=b.style.transition="none";b.style.borderSpacing=b.style.padding="0";si(b,c,"0px");b.style.width=Wf(a)+"px";if(0!==ui(a,b,c)){si(b,c,"0px");var d=ui(a,b,c);si(b,c,-1*d+"px");a=ui(a,b,c);0!==a&&a!==d&&si(b,c,d/(a-d)*d+"px")}b.style.zIndex=30}};function wi(a,b){this.l=a;this.j=b}wi.prototype.minWidth=function(){return this.l};wi.prototype.height=function(){return this.j};wi.prototype.h=function(a){return 300<a&&300<this.j?this.l:Math.min(1200,Math.round(a))};wi.prototype.i=function(){};function xi(a,b,c,d){d=void 0===d?function(f){return f}:d;var e;return a.style&&a.style[c]&&d(a.style[c])||(e=Nc(a,b))&&e[c]&&d(e[c])||null}function yi(a){return function(b){return b.minWidth()<=a}}function zi(a,b,c,d){var e=a&&Ai(c,b),f=Bi(b,d);return function(g){return!(e&&g.height()>=f)}}function Ci(a){return function(b){return b.height()<=a}}function Ai(a,b){return ni(a,b)<Vf(b).clientHeight-100} function Di(a,b){var c=xi(b,a,"height",K);if(c)return c;var d=b.style.height;b.style.height="inherit";c=xi(b,a,"height",K);b.style.height=d;if(c)return c;c=Infinity;do(d=b.style&&K(b.style.height))&&(c=Math.min(c,d)),(d=xi(b,a,"maxHeight",K))&&(c=Math.min(c,d));while((b=b.parentElement)&&"HTML"!=b.tagName);return c}function Bi(a,b){var c=0==pd(a);return b&&c?Math.max(250,2*Vf(a).clientHeight/3):250};var R={},Ei=(R.google_ad_channel=!0,R.google_ad_client=!0,R.google_ad_host=!0,R.google_ad_host_channel=!0,R.google_adtest=!0,R.google_tag_for_child_directed_treatment=!0,R.google_tag_for_under_age_of_consent=!0,R.google_tag_partner=!0,R.google_restrict_data_processing=!0,R.google_page_url=!0,R.google_debug_params=!0,R.google_adbreak_test=!0,R.google_ad_frequency_hint=!0,R.google_admob_interstitial_slot=!0,R.google_admob_rewarded_slot=!0,R.google_max_ad_content_rating=!0,R.google_traffic_source=!0, R),Fi=RegExp("(^| )adsbygoogle($| )");function Gi(a,b){for(var c=0;c<b.length;c++){var d=b[c],e=zc(d.Rb);a[e]=d.value}};function Hi(a,b,c,d){this.l=a;this.i=b;this.j=c;this.h=d}function Ii(a,b){var c=[];try{c=b.querySelectorAll(a.l)}catch(g){}if(!c.length)return[];b=Ya(c);b=Ji(a,b);"number"===typeof a.i&&(c=a.i,0>c&&(c+=b.length),b=0<=c&&c<b.length?[b[c]]:[]);if("number"===typeof a.j){c=[];for(var d=0;d<b.length;d++){var e=Ki(b[d]),f=a.j;0>f&&(f+=e.length);0<=f&&f<e.length&&c.push(e[f])}b=c}return b} Hi.prototype.toString=function(){return JSON.stringify({nativeQuery:this.l,occurrenceIndex:this.i,paragraphIndex:this.j,ignoreMode:this.h})};function Ji(a,b){if(null==a.h)return b;switch(a.h){case 1:return b.slice(1);case 2:return b.slice(0,b.length-1);case 3:return b.slice(1,b.length-1);case 0:return b;default:throw Error("Unknown ignore mode: "+a.h);}}function Ki(a){var b=[];xd(a.getElementsByTagName("p"),function(c){100<=Li(c)&&b.push(c)});return b} function Li(a){if(3==a.nodeType)return a.length;if(1!=a.nodeType||"SCRIPT"==a.tagName)return 0;var b=0;xd(a.childNodes,function(c){b+=Li(c)});return b}function Mi(a){return 0==a.length||isNaN(a[0])?a:"\\"+(30+parseInt(a[0],10))+" "+a.substring(1)};function Ni(a){if(!a)return null;var b=A(a,7);if(A(a,1)||a.getId()||0<wb(a,4).length){var c=a.getId();b=wb(a,4);var d=A(a,1),e="";d&&(e+=d);c&&(e+="#"+Mi(c));if(b)for(c=0;c<b.length;c++)e+="."+Mi(b[c]);a=(b=e)?new Hi(b,A(a,2),A(a,5),Oi(A(a,6))):null}else a=b?new Hi(b,A(a,2),A(a,5),Oi(A(a,6))):null;return a}var Pi={1:1,2:2,3:3,0:0};function Oi(a){return null==a?a:Pi[a]}var Qi={1:0,2:1,3:2,4:3};function Ri(a){return a.google_ama_state=a.google_ama_state||{}} function Si(a){a=Ri(a);return a.optimization=a.optimization||{}};function Ti(a){switch(A(a,8)){case 1:case 2:if(null==a)var b=null;else b=G(a,Jd,1),null==b?b=null:(a=A(a,2),b=null==a?null:new Ld({Ga:[b],Ra:a}));return null!=b?Dd(b):Fd(Error("Missing dimension when creating placement id"));case 3:return Fd(Error("Missing dimension when creating placement id"));default:return Fd(Error("Invalid type: "+A(a,8)))}};function T(a){a=void 0===a?"":a;var b=Error.call(this);this.message=b.message;"stack"in b&&(this.stack=b.stack);this.name="TagError";this.message=a?"adsbygoogle.push() error: "+a:"";Error.captureStackTrace?Error.captureStackTrace(this,T):this.stack=Error().stack||""}v(T,Error);var Pf,Ui,fi=new ci;function Vi(a){null!=a&&(w.google_measure_js_timing=a);w.google_measure_js_timing||(a=fi,a.h=!1,a.i!=a.j.google_js_reporting_queue&&(bi()&&Ra(a.i,di),a.i.length=0))}(function(a){Pf=a||new Nf;"number"!==typeof w.google_srt&&(w.google_srt=Math.random());Of();Ui=new ei;Ui.Va(!0);"complete"==w.document.readyState?Vi():fi.h&&xc(w,"load",function(){Vi()})})();function Wi(a,b,c){return Ui.oa(a,b,c)}function Xi(a,b){return Ui.Oa(a,b)} function Yi(a,b,c){var d=O(Oh).h();!b.eid&&d.length&&(b.eid=d.toString());Qf(Pf,a,b,!0,c)}function Zi(a,b){Ui.Pa(a,b)}function $i(a,b,c,d){var e;Rh(b)?e=b.msg||gi(b.error):e=gi(b);return 0==e.indexOf("TagError")?(c=b instanceof Qh?b.error:b,c.pbr||(c.pbr=!0,Ui.I(a,b,.1,d,"puberror")),!1):Ui.I(a,b,c,d)};function aj(a){a=void 0===a?window:a;a=a.googletag;return(null==a?0:a.apiReady)?a:void 0};function bj(a){var b=aj(a);return b?Sa(Ta(b.pubads().getSlots(),function(c){return a.document.getElementById(c.getSlotElementId())}),function(c){return null!=c}):null}function cj(a,b){return Ya(a.document.querySelectorAll(b))}function dj(a){var b=[];a=u(a);for(var c=a.next();!c.done;c=a.next()){c=c.value;for(var d=!0,e=0;e<b.length;e++){var f=b[e];if(f.contains(c)){d=!1;break}if(c.contains(f)){d=!1;b[e]=c;break}}d&&b.push(c)}return b};function ej(a,b){function c(){d.push({anchor:e.anchor,position:e.position});return e.anchor==b.anchor&&e.position==b.position}for(var d=[],e=a;e;){switch(e.position){case 1:if(c())return d;e.position=2;case 2:if(c())return d;if(e.anchor.firstChild){e={anchor:e.anchor.firstChild,position:1};continue}else e.position=3;case 3:if(c())return d;e.position=4;case 4:if(c())return d}for(;e&&!e.anchor.nextSibling&&e.anchor.parentNode!=e.anchor.ownerDocument.body;){e={anchor:e.anchor.parentNode,position:3}; if(c())return d;e.position=4;if(c())return d}e&&e.anchor.nextSibling?e={anchor:e.anchor.nextSibling,position:1}:e=null}return d};function fj(a,b){this.i=a;this.h=b} function gj(a,b){var c=new Id,d=new Hd;b.forEach(function(e){if(Ib(e,Yd,1,ae)){e=Ib(e,Yd,1,ae);if(G(e,Wd,1)&&G(G(e,Wd,1),Jd,1)&&G(e,Wd,2)&&G(G(e,Wd,2),Jd,1)){var f=hj(a,G(G(e,Wd,1),Jd,1)),g=hj(a,G(G(e,Wd,2),Jd,1));if(f&&g)for(f=u(ej({anchor:f,position:A(G(e,Wd,1),2)},{anchor:g,position:A(G(e,Wd,2),2)})),g=f.next();!g.done;g=f.next())g=g.value,c.set(za(g.anchor),g.position)}G(e,Wd,3)&&G(G(e,Wd,3),Jd,1)&&(f=hj(a,G(G(e,Wd,3),Jd,1)))&&c.set(za(f),A(G(e,Wd,3),2))}else Ib(e,Zd,2,ae)?ij(a,Ib(e,Zd,2,ae), c):Ib(e,$d,3,ae)&&jj(a,Ib(e,$d,3,ae),d)});return new fj(c,d)}function ij(a,b,c){G(b,Jd,1)&&(a=kj(a,G(b,Jd,1)))&&a.forEach(function(d){d=za(d);c.set(d,1);c.set(d,4);c.set(d,2);c.set(d,3)})}function jj(a,b,c){G(b,Jd,1)&&(a=kj(a,G(b,Jd,1)))&&a.forEach(function(d){c.add(za(d))})}function hj(a,b){return(a=kj(a,b))&&0<a.length?a[0]:null}function kj(a,b){return(b=Ni(b))?Ii(b,a):null};function lj(){this.h=new p.Set}function mj(a){a=nj(a);return a.has("all")||a.has("after")}function oj(a){a=nj(a);return a.has("all")||a.has("before")}function pj(a,b,c){switch(c){case 2:case 3:break;case 1:case 4:b=b.parentElement;break;default:throw Error("Unknown RelativePosition: "+c);}for(c=[];b;){if(qj(b))return!0;if(a.h.has(b))break;c.push(b);b=b.parentElement}c.forEach(function(d){return a.h.add(d)});return!1} function qj(a){var b=nj(a);return a&&("AUTO-ADS-EXCLUSION-AREA"===a.tagName||b.has("inside")||b.has("all"))}function nj(a){return(a=a&&a.getAttribute("data-no-auto-ads"))?new p.Set(a.split("|")):new p.Set};function rj(a,b){if(!a)return!1;a=Nc(a,b);if(!a)return!1;a=a.cssFloat||a.styleFloat;return"left"==a||"right"==a}function sj(a){for(a=a.previousSibling;a&&1!=a.nodeType;)a=a.previousSibling;return a?a:null}function tj(a){return!!a.nextSibling||!!a.parentNode&&tj(a.parentNode)};function uj(a){var b={};a&&wb(a,6).forEach(function(c){b[c]=!0});return b}function vj(a,b,c,d,e){this.h=a;this.H=b;this.j=c;this.m=e||null;this.A=(this.C=d)?gj(a.document,H(d,Xd,5)):gj(a.document,[]);this.G=new lj;this.i=0;this.l=!1} function wj(a,b){if(a.l)return!0;a.l=!0;var c=H(a.j,ce,1);a.i=0;var d=uj(a.C);var e=a.h;try{var f=e.localStorage.getItem("google_ama_settings");var g=f?Nb(se,f):null}catch(S){g=null}var h=null!==g&&D(g,2,!1);g=Ri(e);h&&(g.eatf=!0,kd(7,[!0,0,!1]));var k=P(Ve)||P(Ue);f=P(Ue);if(k){b:{var l={fb:!1},m=cj(e,".google-auto-placed"),q=cj(e,'ins.adsbygoogle[data-anchor-shown="true"]'),t=cj(e,"ins.adsbygoogle[data-ad-format=autorelaxed]");var y=(bj(e)||cj(e,"div[id^=div-gpt-ad]")).concat(cj(e,"iframe[id^=google_ads_iframe]")); var F=cj(e,"div.trc_related_container,div.OUTBRAIN,div[id^=rcjsload],div[id^=ligatusframe],div[id^=crt-],iframe[id^=cto_iframe],div[id^=yandex_], div[id^=Ya_sync],iframe[src*=adnxs],div.advertisement--appnexus,div[id^=apn-ad],div[id^=amzn-native-ad],iframe[src*=amazon-adsystem],iframe[id^=ox_],iframe[src*=openx],img[src*=openx],div[class*=adtech],div[id^=adtech],iframe[src*=adtech],div[data-content-ad-placement=true],div.wpcnt div[id^=atatags-]"),z=cj(e,"ins.adsbygoogle-ablated-ad-slot"),E=cj(e,"div.googlepublisherpluginad"); k=[].concat(cj(e,"iframe[id^=aswift_],iframe[id^=google_ads_frame]"),cj(e,"ins.adsbygoogle"));h=[];l=u([[l.Mb,m],[l.fb,q],[l.Pb,t],[l.Nb,y],[l.Qb,F],[l.Lb,z],[l.Ob,E]]);for(m=l.next();!m.done;m=l.next())q=u(m.value),m=q.next().value,q=q.next().value,!1===m?h=h.concat(q):k=k.concat(q);k=dj(k);l=dj(h);h=k.slice(0);k=u(l);for(l=k.next();!l.done;l=k.next())for(l=l.value,m=0;m<h.length;m++)(l.contains(h[m])||h[m].contains(l))&&h.splice(m,1);e=Vf(e).clientHeight;for(k=0;k<h.length;k++)if(l=h[k].getBoundingClientRect(), !(0===l.height&&!f||l.top>e)){e=!0;break b}e=!1}g=e?g.eatfAbg=!0:!1}else g=h;if(g)return!0;g=new Hd([2]);for(e=0;e<c.length;e++){f=a;k=c[e];h=e;l=b;if(!G(k,Pd,4)||!g.contains(A(G(k,Pd,4),1))||1!==A(k,8)||k&&null!=A(k,4)&&d[A(G(k,Pd,4),2)])f=null;else{f.i++;if(k=xj(f,k,l,d))l=Ri(f.h),l.numAutoAdsPlaced||(l.numAutoAdsPlaced=0),null==l.placed&&(l.placed=[]),l.numAutoAdsPlaced++,l.placed.push({index:h,element:k.ha}),kd(7,[!1,f.i,!0]);f=k}if(f)return!0}kd(7,[!1,a.i,!1]);return!1} function xj(a,b,c,d){if(b&&null!=A(b,4)&&d[A(G(b,Pd,4),2)]||1!=A(b,8))return null;d=G(b,Jd,1);if(!d)return null;d=Ni(d);if(!d)return null;d=Ii(d,a.h.document);if(0==d.length)return null;d=d[0];var e=Qi[A(b,2)];e=void 0===e?null:e;var f;if(!(f=null==e)){a:{f=a.h;switch(e){case 0:f=rj(sj(d),f);break a;case 3:f=rj(d,f);break a;case 2:var g=d.lastChild;f=rj(g?1==g.nodeType?g:sj(g):null,f);break a}f=!1}if(c=!f&&!(!c&&2==e&&!tj(d)))c=1==e||2==e?d:d.parentNode,c=!(c&&!te(c)&&0>=c.offsetWidth);f=!c}if(!(c= f)){c=a.A;f=A(b,2);g=za(d);g=c.i.h.get(g);if(!(g=g?g.contains(f):!1))a:{if(c.h.contains(za(d)))switch(f){case 2:case 3:g=!0;break a;default:g=!1;break a}for(f=d.parentElement;f;){if(c.h.contains(za(f))){g=!0;break a}f=f.parentElement}g=!1}c=g}if(!c){c=a.G;f=A(b,2);a:switch(f){case 1:g=mj(d.previousElementSibling)||oj(d);break a;case 4:g=mj(d)||oj(d.nextElementSibling);break a;case 2:g=oj(d.firstElementChild);break a;case 3:g=mj(d.lastElementChild);break a;default:throw Error("Unknown RelativePosition: "+ f);}c=g||pj(c,d,f)}if(c)return null;c=G(b,be,3);f={};c&&(f.Wa=A(c,1),f.Ha=A(c,2),f.cb=!!xb(c,3));c=G(b,Pd,4)&&A(G(b,Pd,4),2)?A(G(b,Pd,4),2):null;c=Sd(c);g=null!=A(b,12)?A(b,12):null;g=null==g?null:new Qd(null,{google_ml_rank:g});b=yj(a,b);b=Rd(a.m,c,g,b);c=a.h;a=a.H;var h=c.document,k=f.cb||!1;g=(new Bc(h)).createElement("DIV");var l=g.style;l.width="100%";l.height="auto";l.clear=k?"both":"none";k=g.style;k.textAlign="center";f.lb&&Gi(k,f.lb);h=(new Bc(h)).createElement("INS");k=h.style;k.display= "block";k.margin="auto";k.backgroundColor="transparent";f.Wa&&(k.marginTop=f.Wa);f.Ha&&(k.marginBottom=f.Ha);f.ab&&Gi(k,f.ab);g.appendChild(h);f={ra:g,ha:h};f.ha.setAttribute("data-ad-format","auto");g=[];if(h=b&&b.Ja)f.ra.className=h.join(" ");h=f.ha;h.className="adsbygoogle";h.setAttribute("data-ad-client",a);g.length&&h.setAttribute("data-ad-channel",g.join("+"));a:{try{var m=f.ra;var q=void 0===q?0:q;if(P(Qe)){q=void 0===q?0:q;var t=Af(d,e,q);if(t.init){var y=t.init;for(d=y;d=t.ja(d);)y=d;var F= {anchor:y,position:t.na}}else F={anchor:d,position:e};m["google-ama-order-assurance"]=q;ue(m,F.anchor,F.position)}else ue(m,d,e);b:{var z=f.ha;z.dataset.adsbygoogleStatus="reserved";z.className+=" adsbygoogle-noablate";m={element:z};var E=b&&b.Qa;if(z.hasAttribute("data-pub-vars")){try{E=JSON.parse(z.getAttribute("data-pub-vars"))}catch(S){break b}z.removeAttribute("data-pub-vars")}E&&(m.params=E);(c.adsbygoogle=c.adsbygoogle||[]).push(m)}}catch(S){(z=f.ra)&&z.parentNode&&(E=z.parentNode,E.removeChild(z), te(E)&&(E.style.display=E.getAttribute("data-init-display")||"none"));z=!1;break a}z=!0}return z?f:null}function yj(a,b){return Bd(Ed(Ti(b).map(Td),function(c){Ri(a.h).exception=c}))};function zj(a){if(P(Pe))var b=null;else try{b=a.getItem("google_ama_config")}catch(d){b=null}try{var c=b?Nb(je,b):null}catch(d){c=null}return c};function Aj(a){J.call(this,a)}v(Aj,J);function Bj(a){try{var b=a.localStorage.getItem("google_auto_fc_cmp_setting")||null}catch(d){b=null}var c=b;return c?Gd(function(){return Nb(Aj,c)}):Dd(null)};function Cj(){this.S={}}function Dj(){if(Ej)return Ej;var a=md()||window,b=a.google_persistent_state_async;return null!=b&&"object"==typeof b&&null!=b.S&&"object"==typeof b.S?Ej=b:a.google_persistent_state_async=Ej=new Cj}function Fj(a){return Gj[a]||"google_ps_"+a}function Hj(a,b,c){b=Fj(b);a=a.S;var d=a[b];return void 0===d?a[b]=c:d}var Ej=null,Ij={},Gj=(Ij[8]="google_prev_ad_formats_by_region",Ij[9]="google_prev_ad_slotnames_by_region",Ij);function Jj(a){this.h=a||{cookie:""}} Jj.prototype.set=function(a,b,c){var d=!1;if("object"===typeof c){var e=c.Sb;d=c.Tb||!1;var f=c.domain||void 0;var g=c.path||void 0;var h=c.jb}if(/[;=\s]/.test(a))throw Error('Invalid cookie name "'+a+'"');if(/[;\r\n]/.test(b))throw Error('Invalid cookie value "'+b+'"');void 0===h&&(h=-1);this.h.cookie=a+"="+b+(f?";domain="+f:"")+(g?";path="+g:"")+(0>h?"":0==h?";expires="+(new Date(1970,1,1)).toUTCString():";expires="+(new Date(Date.now()+1E3*h)).toUTCString())+(d?";secure":"")+(null!=e?";samesite="+ e:"")};Jj.prototype.get=function(a,b){for(var c=a+"=",d=(this.h.cookie||"").split(";"),e=0,f;e<d.length;e++){f=Ja(d[e]);if(0==f.lastIndexOf(c,0))return f.substr(c.length);if(f==a)return""}return b};Jj.prototype.isEmpty=function(){return!this.h.cookie}; Jj.prototype.clear=function(){for(var a=(this.h.cookie||"").split(";"),b=[],c=[],d,e,f=0;f<a.length;f++)e=Ja(a[f]),d=e.indexOf("="),-1==d?(b.push(""),c.push(e)):(b.push(e.substring(0,d)),c.push(e.substring(d+1)));for(a=b.length-1;0<=a;a--)c=b[a],this.get(c),this.set(c,"",{jb:0,path:void 0,domain:void 0})};function Kj(a){J.call(this,a)}v(Kj,J);function Lj(a){var b=new Kj;return B(b,5,a)};function Mj(){this.A=this.A;this.G=this.G}Mj.prototype.A=!1;Mj.prototype.j=function(){if(this.G)for(;this.G.length;)this.G.shift()()};function Nj(a){void 0!==a.addtlConsent&&"string"!==typeof a.addtlConsent&&(a.addtlConsent=void 0);void 0!==a.gdprApplies&&"boolean"!==typeof a.gdprApplies&&(a.gdprApplies=void 0);return void 0!==a.tcString&&"string"!==typeof a.tcString||void 0!==a.listenerId&&"number"!==typeof a.listenerId?2:a.cmpStatus&&"error"!==a.cmpStatus?0:3}function Oj(a,b){b=void 0===b?500:b;Mj.call(this);this.h=a;this.i=null;this.m={};this.H=0;this.C=b;this.l=null}v(Oj,Mj); Oj.prototype.j=function(){this.m={};this.l&&(yc(this.h,this.l),delete this.l);delete this.m;delete this.h;delete this.i;Mj.prototype.j.call(this)};function Pj(a){return"function"===typeof a.h.__tcfapi||null!=Qj(a)} Oj.prototype.addEventListener=function(a){function b(f,g){clearTimeout(e);f?(c=f,c.internalErrorState=Nj(c),g&&0===c.internalErrorState||(c.tcString="tcunavailable",g||(c.internalErrorState=3))):(c.tcString="tcunavailable",c.internalErrorState=3);a(c)}var c={},d=wc(function(){return a(c)}),e=0;-1!==this.C&&(e=setTimeout(function(){c.tcString="tcunavailable";c.internalErrorState=1;d()},this.C));try{Rj(this,"addEventListener",b)}catch(f){c.tcString="tcunavailable",c.internalErrorState=3,e&&(clearTimeout(e), e=0),d()}};Oj.prototype.removeEventListener=function(a){a&&a.listenerId&&Rj(this,"removeEventListener",null,a.listenerId)};function Rj(a,b,c,d){c||(c=function(){});if("function"===typeof a.h.__tcfapi)a=a.h.__tcfapi,a(b,2,c,d);else if(Qj(a)){Sj(a);var e=++a.H;a.m[e]=c;a.i&&(c={},a.i.postMessage((c.__tcfapiCall={command:b,version:2,callId:e,parameter:d},c),"*"))}else c({},!1)}function Qj(a){if(a.i)return a.i;a.i=$c(a.h,"__tcfapiLocator");return a.i} function Sj(a){a.l||(a.l=function(b){try{var c=("string"===typeof b.data?JSON.parse(b.data):b.data).__tcfapiReturn;a.m[c.callId](c.returnValue,c.success)}catch(d){}},xc(a.h,"message",a.l))};function Tj(a){var b=a.u,c=a.ta,d=a.Ia;a=Uj({u:b,Z:a.Z,ka:void 0===a.ka?!1:a.ka,la:void 0===a.la?!1:a.la});null!=a.h||"tcunav"!=a.i.message?d(a):Vj(b,c).then(function(e){return e.map(Wj)}).then(function(e){return e.map(function(f){return Xj(b,f)})}).then(d)} function Uj(a){var b=a.u,c=a.Z,d=void 0===a.ka?!1:a.ka;if(!(a=!(void 0===a.la?0:a.la)&&Pj(new Oj(b)))){if(d=!d){if(c){c=Bj(b);if(null==c.h)Ui.I(806,c.i,void 0,void 0),c=!1;else if((c=c.h.value)&&null!=A(c,1))b:switch(c=A(c,1),c){case 1:c=!0;break b;default:throw Error("Unhandled AutoGdprFeatureStatus: "+c);}else c=!1;c=!c}d=c}a=d}if(!a)return Xj(b,Lj(!0));c=Dj();return(c=Hj(c,24))?Xj(b,Wj(c)):Fd(Error("tcunav"))}function Vj(a,b){return p.Promise.race([Yj(),Zj(a,b)])} function Yj(){return(new p.Promise(function(a){var b=Dj();a={resolve:a};var c=Hj(b,25,[]);c.push(a);b.S[Fj(25)]=c})).then(ak)}function Zj(a,b){return new p.Promise(function(c){a.setTimeout(c,b,Fd(Error("tcto")))})}function ak(a){return a?Dd(a):Fd(Error("tcnull"))} function Wj(a){var b=void 0===b?!1:b;if(!1===a.gdprApplies)var c=!0;else void 0===a.internalErrorState&&(a.internalErrorState=Nj(a)),c="error"===a.cmpStatus||0!==a.internalErrorState||"loaded"===a.cmpStatus&&("tcloaded"===a.eventStatus||"useractioncomplete"===a.eventStatus)?!0:!1;if(c)if(!1===a.gdprApplies||"tcunavailable"===a.tcString||void 0===a.gdprApplies&&!b||"string"!==typeof a.tcString||!a.tcString.length)a=!0;else{var d=void 0===d?"755":d;b:{if(a.publisher&&a.publisher.restrictions&&(b=a.publisher.restrictions["1"], void 0!==b)){b=b[void 0===d?"755":d];break b}b=void 0}0===b?a=!1:a.purpose&&a.vendor?(b=a.vendor.consents,(d=!(!b||!b[void 0===d?"755":d]))&&a.purposeOneTreatment&&"CH"===a.publisherCC?a=!0:(d&&(a=a.purpose.consents,d=!(!a||!a["1"])),a=d)):a=!0}else a=!1;return Lj(a)}function Xj(a,b){a:{a=void 0===a?window:a;if(xb(b,5))try{var c=a.localStorage;break a}catch(d){}c=null}return(b=c)?Dd(b):Fd(Error("unav"))};function bk(a){J.call(this,a)}v(bk,J);function ck(a){J.call(this,a,-1,dk)}v(ck,J);var dk=[1,2];function ek(a){this.exception=a}function fk(a,b,c){this.j=a;this.h=b;this.i=c}fk.prototype.start=function(){this.l()};fk.prototype.l=function(){try{switch(this.j.document.readyState){case "complete":case "interactive":wj(this.h,!0);gk(this);break;default:wj(this.h,!1)?gk(this):this.j.setTimeout(Ea(this.l,this),100)}}catch(a){gk(this,a)}};function gk(a,b){try{var c=a.i,d=c.resolve,e=a.h;Ri(e.h);H(e.j,ce,1);d.call(c,new ek(b))}catch(f){a.i.reject(f)}};function hk(a){J.call(this,a,-1,ik)}v(hk,J);function jk(a){J.call(this,a)}v(jk,J);function kk(a){J.call(this,a)}v(kk,J);var ik=[7];function lk(a){a=(a=(new Jj(a)).get("FCCDCF",""))?a:null;try{return a?Nb(hk,a):null}catch(b){return null}};Zb({Gb:0,Fb:1,Cb:2,xb:3,Db:4,yb:5,Eb:6,Ab:7,Bb:8,wb:9,zb:10}).map(function(a){return Number(a)});Zb({Ib:0,Jb:1,Hb:2}).map(function(a){return Number(a)});function mk(a){function b(){if(!a.frames.__uspapiLocator)if(c.body){var d=Mc("IFRAME",c);d.style.display="none";d.style.width="0px";d.style.height="0px";d.style.border="none";d.style.zIndex="-1000";d.style.left="-1000px";d.style.top="-1000px";d.name="__uspapiLocator";c.body.appendChild(d)}else a.setTimeout(b,5)}var c=a.document;b()};function nk(a){this.h=a;this.i=a.document;this.j=(a=(a=lk(this.i))?G(a,kk,5)||null:null)?A(a,2):null;(a=lk(this.i))&&G(a,jk,4);(a=lk(this.i))&&G(a,jk,4)}function ok(){var a=window;a.__uspapi||a.frames.__uspapiLocator||(a=new nk(a),pk(a))}function pk(a){!a.j||a.h.__uspapi||a.h.frames.__uspapiLocator||(a.h.__uspapiManager="fc",mk(a.h),Ga(function(){return a.l.apply(a,ka(ta.apply(0,arguments)))}))} nk.prototype.l=function(a,b,c){"function"===typeof c&&"getUSPData"===a&&c({version:1,uspString:this.j},!0)};function qk(a){J.call(this,a)}v(qk,J);qk.prototype.getWidth=function(){return C(this,1,0)};qk.prototype.getHeight=function(){return C(this,2,0)};function rk(a){J.call(this,a)}v(rk,J);function sk(a){J.call(this,a)}v(sk,J);var tk=[4,5];function uk(a){var b=/[a-zA-Z0-9._~-]/,c=/%[89a-zA-Z]./;return a.replace(/(%[a-zA-Z0-9]{2})/g,function(d){if(!d.match(c)){var e=decodeURIComponent(d);if(e.match(b))return e}return d.toUpperCase()})}function vk(a){for(var b="",c=/[/%?&=]/,d=0;d<a.length;++d){var e=a[d];b=e.match(c)?b+e:b+encodeURIComponent(e)}return b};function wk(a,b){a=vk(uk(a.location.pathname)).replace(/(^\/)|(\/$)/g,"");var c=Tc(a),d=xk(a);return r(b,"find").call(b,function(e){var f=null!=A(e,7)?A(G(e,oe,7),1):A(e,1);e=null!=A(e,7)?A(G(e,oe,7),2):2;if("number"!==typeof f)return!1;switch(e){case 1:return f==c;case 2:return d[f]||!1}return!1})||null}function xk(a){for(var b={};;){b[Tc(a)]=!0;if(!a)return b;a=a.substring(0,a.lastIndexOf("/"))}};var yk={},zk=(yk.google_ad_channel=!0,yk.google_ad_host=!0,yk);function Ak(a,b){a.location.href&&a.location.href.substring&&(b.url=a.location.href.substring(0,200));Yi("ama",b,.01)}function Bk(a){var b={};Sc(zk,function(c,d){d in a&&(b[d]=a[d])});return b};function Ck(a){a=G(a,le,3);return!a||A(a,1)<=Date.now()?!1:!0}function Dk(a){return(a=zj(a))?Ck(a)?a:null:null}function Ek(a,b){try{b.removeItem("google_ama_config")}catch(c){Ak(a,{lserr:1})}};function Fk(a){J.call(this,a)}v(Fk,J);function Gk(a){J.call(this,a,-1,Hk)}v(Gk,J);var Hk=[1];function Ik(a){J.call(this,a,-1,Jk)}v(Ik,J);Ik.prototype.getId=function(){return C(this,1,0)};Ik.prototype.V=function(){return C(this,7,0)};var Jk=[2];function Kk(a){J.call(this,a,-1,Lk)}v(Kk,J);Kk.prototype.V=function(){return C(this,5,0)};var Lk=[2];function Mk(a){J.call(this,a,-1,Nk)}v(Mk,J);function Ok(a){J.call(this,a,-1,Pk)}v(Ok,J);Ok.prototype.V=function(){return C(this,1,0)};function Qk(a){J.call(this,a)}v(Qk,J);var Nk=[1,4,2,3],Pk=[2];function Rk(a){J.call(this,a,-1,Sk)}v(Rk,J);function Tk(a){return Ib(a,Gk,14,Uk)}var Sk=[19],Uk=[13,14];var Vk=void 0;function Wk(){Yf(Vk,Xf);return Vk}function Xk(a){Yf(Vk,$f);Vk=a};function Yk(a,b,c,d){c=void 0===c?"":c;return 1===b&&Zk(c,void 0===d?null:d)?!0:$k(a,c,function(e){return Ua(H(e,Tb,2),function(f){return A(f,1)===b})})}function Zk(a,b){return b?13===Cb(b,Uk)?D(Ib(b,Fk,13,Uk),1):14===Cb(b,Uk)&&""!==a&&1===wb(Tk(b),1).length&&wb(Tk(b),1)[0]===a?D(G(Tk(b),Fk,2),1):!1:!1}function al(a,b){b=C(b,18,0);-1!==b&&(a.tmod=b)}function bl(a){var b=void 0===b?"":b;var c=Ic(L)||L;return cl(c,a)?!0:$k(L,b,function(d){return Ua(wb(d,3),function(e){return e===a})})} function dl(a){return $k(w,void 0===a?"":a,function(){return!0})}function cl(a,b){a=(a=(a=a.location&&a.location.hash)&&a.match(/forced_clientside_labs=([\d,]+)/))&&a[1];return!!a&&Xa(a.split(","),b.toString())}function $k(a,b,c){a=Ic(a)||a;var d=el(a);b&&(b=rd(String(b)));return Yb(d,function(e,f){return Object.prototype.hasOwnProperty.call(d,f)&&(!b||b===f)&&c(e)})}function el(a){a=fl(a);var b={};Sc(a,function(c,d){try{var e=new Rb(c);b[d]=e}catch(f){}});return b} function fl(a){return P(xe)?(a=Uj({u:a,Z:Wk()}),null!=a.h?(gl("ok"),a=hl(a.h.value)):(gl(a.i.message),a={}),a):hl(a.localStorage)}function hl(a){try{var b=a.getItem("google_adsense_settings");if(!b)return{};var c=JSON.parse(b);return c!==Object(c)?{}:Xb(c,function(d,e){return Object.prototype.hasOwnProperty.call(c,e)&&"string"===typeof e&&Array.isArray(d)})}catch(d){return{}}}function gl(a){P(we)&&Yi("abg_adsensesettings_lserr",{s:a,g:P(xe),c:Wk(),r:.01},.01)};function il(a,b,c,d){jl(new kl(a,b,c,d))}function kl(a,b,c,d){this.u=a;this.i=b;this.j=c;this.h=d}function jl(a){Ed(Cd(Uj({u:a.u,Z:D(a.i,6)}),function(b){ll(a,b,!0)}),function(){ml(a)})}function ll(a,b,c){Ed(Cd(nl(b),function(d){ol("ok");a.h(d)}),function(){Ek(a.u,b);c?ml(a):a.h(null)})}function ml(a){Ed(Cd(pl(a),a.h),function(){ql(a)})}function ql(a){Tj({u:a.u,Z:D(a.i,6),ta:50,Ia:function(b){rl(a,b)}})}function nl(a){return(a=Dk(a))?Dd(a):Fd(Error("invlocst"))} function pl(a){a:{var b=a.u;var c=a.j;a=a.i;if(13===Cb(a,Uk))b=(b=G(G(Ib(a,Fk,13,Uk),bk,2),ck,2))&&0<H(b,ce,1).length?b:null;else{if(14===Cb(a,Uk)){var d=wb(Tk(a),1),e=G(G(G(Tk(a),Fk,2),bk,2),ck,2);if(1===d.length&&d[0]===c&&e&&0<H(e,ce,1).length&&I(a,17)===b.location.host){b=e;break a}}b=null}}b?(c=new je,a=H(b,ce,1),c=Gb(c,1,a),b=H(b,me,2),b=Gb(c,7,b),b=Dd(b)):b=Fd(Error("invtag"));return b}function rl(a,b){Ed(Cd(b,function(c){ll(a,c,!1)}),function(c){ol(c.message);a.h(null)})} function ol(a){Yi("abg::amalserr",{status:a,guarding:"true",timeout:50,rate:.01},.01)};function sl(a){Ak(a,{atf:1})}function tl(a,b){(a.google_ama_state=a.google_ama_state||{}).exception=b;Ak(a,{atf:0})};function U(a){a.google_ad_modifications||(a.google_ad_modifications={});return a.google_ad_modifications}function ul(a){a=U(a);var b=a.space_collapsing||"none";return a.remove_ads_by_default?{Fa:!0,tb:b,qa:a.ablation_viewport_offset}:null}function vl(a,b){a=U(a);a.had_ads_ablation=!0;a.remove_ads_by_default=!0;a.space_collapsing="slot";a.ablation_viewport_offset=b}function wl(a){U(L).allow_second_reactive_tag=a} function xl(){var a=U(window);a.afg_slotcar_vars||(a.afg_slotcar_vars={});return a.afg_slotcar_vars};function yl(a,b){if(!a)return!1;a=a.hash;if(!a||!a.indexOf)return!1;if(-1!=a.indexOf(b))return!0;b=zl(b);return"go"!=b&&-1!=a.indexOf(b)?!0:!1}function zl(a){var b="";Sc(a.split("_"),function(c){b+=c.substr(0,2)});return b};$a||!x("Safari")||Oa();function Al(){var a=this;this.promise=new p.Promise(function(b,c){a.resolve=b;a.reject=c})};function Bl(){var a=new Al;return{promise:a.promise,resolve:a.resolve}};function Cl(a){a=void 0===a?function(){}:a;w.google_llp||(w.google_llp={});var b=w.google_llp,c=b[7];if(c)return c;c=Bl();b[7]=c;a();return c}function Dl(a){return Cl(function(){Lc(w.document,a)}).promise};function El(a){var b={};return{enable_page_level_ads:(b.pltais=!0,b),google_ad_client:a}};function Fl(a){if(w.google_apltlad||w!==w.top||!a.google_ad_client)return null;w.google_apltlad=!0;var b=El(a.google_ad_client),c=b.enable_page_level_ads;Sc(a,function(d,e){Ei[e]&&"google_ad_client"!==e&&(c[e]=d)});c.google_pgb_reactive=7;if("google_ad_section"in a||"google_ad_region"in a)c.google_ad_section=a.google_ad_section||a.google_ad_region;return b}function Gl(a){return ya(a.enable_page_level_ads)&&7===a.enable_page_level_ads.google_pgb_reactive};function Hl(a,b){this.h=w;this.i=a;this.j=b}function Il(a){P(lf)?il(a.h,a.j,a.i.google_ad_client||"",function(b){var c=a.h,d=a.i;U(L).ama_ran_on_page||b&&Jl(c,d,b)}):Tj({u:a.h,Z:D(a.j,6),ta:50,Ia:function(b){return Kl(a,b)}})}function Kl(a,b){Ed(Cd(b,function(c){Ll("ok");var d=a.h,e=a.i;if(!U(L).ama_ran_on_page){var f=Dk(c);f?Jl(d,e,f):Ek(d,c)}}),function(c){return Ll(c.message)})}function Ll(a){Yi("abg::amalserr",{status:a,guarding:!0,timeout:50,rate:.01},.01)} function Jl(a,b,c){if(null!=A(c,24)){var d=Si(a);d.availableAbg=!0;var e,f;d.ablationFromStorage=!!(null==(e=G(c,ee,24))?0:null==(f=G(e,ge,3))?0:Ib(f,he,2,ie))}if(Gl(b)&&(d=wk(a,H(c,me,7)),!d||!xb(d,8)))return;U(L).ama_ran_on_page=!0;var g;if(null==(g=G(c,re,15))?0:xb(g,23))U(a).enable_overlap_observer=!0;if((g=G(c,pe,13))&&1===A(g,1)){var h=0,k=G(g,qe,6);k&&A(k,3)&&(h=A(k,3)||0);vl(a,h)}else if(null==(h=G(c,ee,24))?0:null==(k=G(h,ge,3))?0:Ib(k,he,2,ie))Si(a).ablatingThisPageview=!0,vl(a,1);kd(3, [c.toJSON()]);var l=b.google_ad_client||"";b=Bk(ya(b.enable_page_level_ads)?b.enable_page_level_ads:{});var m=Rd(Vd,new Qd(null,b));Wi(782,function(){var q=m;try{var t=wk(a,H(c,me,7)),y;if(y=t)a:{var F=wb(t,2);if(F)for(var z=0;z<F.length;z++)if(1==F[z]){y=!0;break a}y=!1}if(y){if(A(t,4)){y={};var E=new Qd(null,(y.google_package=A(t,4),y));q=Rd(q,E)}var S=new vj(a,l,c,t,q),rb=new sd;(new fk(a,S,rb)).start();rb.i.then(Fa(sl,a),Fa(tl,a))}}catch(Kc){Ak(a,{atf:-1})}})};/* Copyright The Closure Library Authors. SPDX-License-Identifier: Apache-2.0 */ var Ml=ja(["https://fonts.googleapis.com/css2?family=Google+Material+Icons:wght@400;500;700"]);function Nl(a,b){return a instanceof HTMLScriptElement&&b.test(a.src)?0:1}function Ol(a){var b=L.document;if(b.currentScript)return Nl(b.currentScript,a);b=u(b.scripts);for(var c=b.next();!c.done;c=b.next())if(0===Nl(c.value,a))return 0;return 1};function Pl(a,b){var c={},d={},e={},f={};return f[fg]=(c[55]=function(){return 0===a},c[23]=function(g){return Yk(L,Number(g))},c[24]=function(g){return bl(Number(g))},c[61]=function(){return D(b,6)},c[63]=function(){return D(b,6)||".google.ch"===I(b,8)},c),f[gg]=(d[7]=function(g){try{var h=window.localStorage}catch(l){h=null}g=Number(g);g=void 0===g?0:g;g=0!==g?"google_experiment_mod"+g:"google_experiment_mod";var k=Vc(h,g);h=null===k?Wc(h,g):k;return null!=h?h:void 0},d),f[hg]=(e[6]=function(){return I(b, 15)},e),f};function Ql(a){a=void 0===a?w:a;return a.ggeac||(a.ggeac={})};function Rl(a,b){try{var c=a.split(".");a=w;for(var d=0,e;null!=a&&d<c.length;d++)e=a,a=a[c[d]],"function"===typeof a&&(a=e[c[d]]());var f=a;if(typeof f===b)return f}catch(g){}} function Sl(){var a={};this[fg]=(a[8]=function(b){try{return null!=va(b)}catch(c){}},a[9]=function(b){try{var c=va(b)}catch(d){return}if(b="function"===typeof c)c=c&&c.toString&&c.toString(),b="string"===typeof c&&-1!=c.indexOf("[native code]");return b},a[10]=function(){return window==window.top},a[6]=function(b){return Xa(O(Oh).h(),parseInt(b,10))},a[27]=function(b){b=Rl(b,"boolean");return void 0!==b?b:void 0},a[60]=function(b){try{return!!w.document.querySelector(b)}catch(c){}},a);a={};this[gg]= (a[3]=function(){return ad()},a[6]=function(b){b=Rl(b,"number");return void 0!==b?b:void 0},a[11]=function(b){b=void 0===b?"":b;var c=w;b=void 0===b?"":b;c=void 0===c?window:c;b=(c=(c=c.location.href.match(Ec)[3]||null)?decodeURI(c):c)?Tc(c+b):null;return null==b?void 0:b%1E3},a);a={};this[hg]=(a[2]=function(){return window.location.href},a[3]=function(){try{return window.top.location.hash}catch(b){return""}},a[4]=function(b){b=Rl(b,"string");return void 0!==b?b:void 0},a[10]=function(){try{var b= w.document;return b.visibilityState||b.webkitVisibilityState||b.mozVisibilityState||""}catch(c){return""}},a[11]=function(){try{var b,c,d,e,f;return null!=(f=null==(d=null==(b=va("google_tag_data"))?void 0:null==(c=b.uach)?void 0:c.fullVersionList)?void 0:null==(e=r(d,"find").call(d,function(g){return"Google Chrome"===g.brand}))?void 0:e.version)?f:""}catch(g){return""}},a)};var Tl=[12,13,20];function Ul(){}Ul.prototype.init=function(a,b,c,d){var e=this;d=void 0===d?{}:d;var f=void 0===d.Ka?!1:d.Ka,g=void 0===d.kb?{}:d.kb;d=void 0===d.mb?[]:d.mb;this.l=a;this.A={};this.G=f;this.m=g;a={};this.i=(a[b]=[],a[4]=[],a);this.j={};(b=Wh())&&Ra(b.split(",")||[],function(h){(h=parseInt(h,10))&&(e.j[h]=!0)});Ra(d,function(h){e.j[h]=!0});this.h=c;return this}; function Vl(a,b,c){var d=[],e=Wl(a.l,b),f;if(f=9!==b)a.A[b]?f=!0:(a.A[b]=!0,f=!1);if(f){var g;null==(g=a.h)||Yg(g,b,c,d,[],4);return d}if(!e.length){var h;null==(h=a.h)||Yg(h,b,c,d,[],3);return d}var k=Xa(Tl,b),l=[];Ra(e,function(q){var t=new Jg;if(q=Xl(a,q,c,t))0!==Cb(t,Kg)&&l.push(t),t=q.getId(),d.push(t),Yl(a,t,k?4:c),(q=H(q,qg,2))&&(k?oh(q,qh(),a.h,t):oh(q,[c],a.h,t))});var m;null==(m=a.h)||Yg(m,b,c,d,l,1);return d}function Yl(a,b,c){a.i[c]||(a.i[c]=[]);a=a.i[c];Xa(a,b)||a.push(b)} function Zl(a,b){a.l.push.apply(a.l,ka(Sa(Ta(b,function(c){return new Ok(c)}),function(c){return!Xa(Tl,c.V())})))} function Xl(a,b,c,d){var e=O(ah).h;if(!mg(G(b,ag,3),e))return null;var f=H(b,Ik,2),g=C(b,6,0);if(g){Bb(d,1,Kg,g);f=e[gg];switch(c){case 2:var h=f[8];break;case 1:h=f[7]}c=void 0;if(h)try{c=h(g),Ab(d,3,c)}catch(k){}return(b=$l(b,c))?am(a,[b],1):null}if(g=C(b,10,0)){Bb(d,2,Kg,g);h=null;switch(c){case 1:h=e[gg][9];break;case 2:h=e[gg][10];break;default:return null}c=h?h(String(g)):void 0;if(void 0===c&&1===C(b,11,0))return null;void 0!==c&&Ab(d,3,c);return(b=$l(b,c))?am(a,[b],1):null}d=e?Sa(f,function(k){return mg(G(k, ag,3),e)}):f;if(!d.length)return null;c=d.length*C(b,1,0);return(b=C(b,4,0))?bm(a,b,c,d):am(a,d,c/1E3)}function bm(a,b,c,d){var e=null!=a.m[b]?a.m[b]:1E3;if(0>=e)return null;d=am(a,d,c/e);a.m[b]=d?0:e-c;return d}function am(a,b,c){var d=a.j,e=Va(b,function(f){return!!d[f.getId()]});return e?e:a.G?null:Oc(b,c)} function cm(a,b){Jh(th,function(c){a.j[c]=!0},b);Jh(wh,function(c,d){return Vl(a,c,d)},b);Jh(xh,function(c){return(a.i[c]||[]).concat(a.i[4])},b);Jh(Gh,function(c){return Zl(a,c)},b);Jh(uh,function(c,d){return Yl(a,c,d)},b)}function Wl(a,b){return(a=Va(a,function(c){return c.V()==b}))&&H(a,Kk,2)||[]}function $l(a,b){var c=H(a,Ik,2),d=c.length,e=C(a,8,0);a=d*C(a,1,0)-1;b=void 0!==b?b:Math.floor(1E3*Rc());d=(b-e)%d;if(b<e||b-e-d>=a)return null;c=c[d];e=O(ah).h;return!c||e&&!mg(G(c,ag,3),e)?null:c};function dm(){this.h=function(){}}function em(a){O(dm).h(a)};var fm,gm,hm,im,jm,km; function lm(a,b,c,d){var e=1;d=void 0===d?Ql():d;e=void 0===e?0:e;var f=void 0===f?new Tg(null!=(im=null==(fm=G(a,Qk,5))?void 0:C(fm,2,0))?im:0,null!=(jm=null==(gm=G(a,Qk,5))?void 0:C(gm,4,0))?jm:0,null!=(km=null==(hm=G(a,Qk,5))?void 0:D(hm,3))?km:!1):f;d.hasOwnProperty("init-done")?(Kh(Gh,d)(Ta(H(a,Ok,2),function(g){return g.toJSON()})),Kh(Hh,d)(Ta(H(a,qg,1),function(g){return g.toJSON()}),e),b&&Kh(Ih,d)(b),mm(d,e)):(cm(O(Ul).init(H(a,Ok,2),e,f,c),d),Lh(d),Mh(d),Nh(d),mm(d,e),oh(H(a,qg,1),[e],f, void 0,!0),bh=bh||!(!c||!c.hb),em(O(Sl)),b&&em(b))}function mm(a,b){a=void 0===a?Ql():a;b=void 0===b?0:b;var c=a,d=b;d=void 0===d?0:d;Ph(O(Oh),c,d);nm(a,b);O(dm).h=Kh(Ih,a);O(yf).m()}function nm(a,b){var c=O(yf);c.i=function(d,e){return Kh(zh,a,function(){return!1})(d,e,b)};c.j=function(d,e){return Kh(Ah,a,function(){return 0})(d,e,b)};c.l=function(d,e){return Kh(Bh,a,function(){return""})(d,e,b)};c.h=function(d,e){return Kh(Ch,a,function(){return[]})(d,e,b)};c.m=function(){Kh(vh,a)(b)}};function om(a,b,c){var d=U(a);if(d.plle)mm(Ql(a),1);else{d.plle=!0;try{var e=a.localStorage}catch(f){e=null}d=e;null==Vc(d,"goog_pem_mod")&&Wc(d,"goog_pem_mod");d=G(b,Mk,12);e=D(b,9);lm(d,Pl(c,b),{Ka:e&&!!a.google_disable_experiments,hb:e},Ql(a));if(c=I(b,15))c=Number(c),O(Oh).l(c);if(c=I(b,10))c=Number(c),O(Oh).i(c);b=u(wb(b,19));for(c=b.next();!c.done;c=b.next())c=c.value,O(Oh).i(c);O(Oh).j(12);O(Oh).j(10);a=Ic(a)||a;yl(a.location,"google_mc_lab")&&O(Oh).i(44738307)}};function pm(a,b,c){a=a.style;a.border="none";a.height=c+"px";a.width=b+"px";a.margin=0;a.padding=0;a.position="relative";a.visibility="visible";a.backgroundColor="transparent"};var qm={"120x90":!0,"160x90":!0,"180x90":!0,"200x90":!0,"468x15":!0,"728x15":!0};function rm(a,b){if(15==b){if(728<=a)return 728;if(468<=a)return 468}else if(90==b){if(200<=a)return 200;if(180<=a)return 180;if(160<=a)return 160;if(120<=a)return 120}return null};function V(a,b,c,d){d=void 0===d?!1:d;wi.call(this,a,b);this.da=c;this.ib=d}v(V,wi);V.prototype.pa=function(){return this.da};V.prototype.i=function(a,b,c){b.google_ad_resize||(c.style.height=this.height()+"px",b.rpe=!0)};function sm(a){return function(b){return!!(b.da&a)}};var tm={},um=(tm.image_stacked=1/1.91,tm.image_sidebyside=1/3.82,tm.mobile_banner_image_sidebyside=1/3.82,tm.pub_control_image_stacked=1/1.91,tm.pub_control_image_sidebyside=1/3.82,tm.pub_control_image_card_stacked=1/1.91,tm.pub_control_image_card_sidebyside=1/3.74,tm.pub_control_text=0,tm.pub_control_text_card=0,tm),vm={},wm=(vm.image_stacked=80,vm.image_sidebyside=0,vm.mobile_banner_image_sidebyside=0,vm.pub_control_image_stacked=80,vm.pub_control_image_sidebyside=0,vm.pub_control_image_card_stacked= 85,vm.pub_control_image_card_sidebyside=0,vm.pub_control_text=80,vm.pub_control_text_card=80,vm),xm={},ym=(xm.pub_control_image_stacked=100,xm.pub_control_image_sidebyside=200,xm.pub_control_image_card_stacked=150,xm.pub_control_image_card_sidebyside=250,xm.pub_control_text=100,xm.pub_control_text_card=150,xm); function zm(a){var b=0;a.T&&b++;a.J&&b++;a.K&&b++;if(3>b)return{M:"Tags data-matched-content-ui-type, data-matched-content-columns-num and data-matched-content-rows-num should be set together."};b=a.T.split(",");var c=a.K.split(",");a=a.J.split(",");if(b.length!==c.length||b.length!==a.length)return{M:'Lengths of parameters data-matched-content-ui-type, data-matched-content-columns-num and data-matched-content-rows-num must match. Example: \n data-matched-content-rows-num="4,2"\ndata-matched-content-columns-num="1,6"\ndata-matched-content-ui-type="image_stacked,image_card_sidebyside"'}; if(2<b.length)return{M:"The parameter length of attribute data-matched-content-ui-type, data-matched-content-columns-num and data-matched-content-rows-num is too long. At most 2 parameters for each attribute are needed: one for mobile and one for desktop, while you are providing "+(b.length+' parameters. Example: \n data-matched-content-rows-num="4,2"\ndata-matched-content-columns-num="1,6"\ndata-matched-content-ui-type="image_stacked,image_card_sidebyside".')};for(var d=[],e=[],f=0;f<b.length;f++){var g= Number(c[f]);if(r(Number,"isNaN").call(Number,g)||0===g)return{M:"Wrong value '"+c[f]+"' for data-matched-content-rows-num."};d.push(g);g=Number(a[f]);if(r(Number,"isNaN").call(Number,g)||0===g)return{M:"Wrong value '"+a[f]+"' for data-matched-content-columns-num."};e.push(g)}return{K:d,J:e,Na:b}} function Am(a){return 1200<=a?{width:1200,height:600}:850<=a?{width:a,height:Math.floor(.5*a)}:550<=a?{width:a,height:Math.floor(.6*a)}:468<=a?{width:a,height:Math.floor(.7*a)}:{width:a,height:Math.floor(3.44*a)}};var Bm=Za("script");function Cm(a,b,c,d,e,f,g,h,k,l,m,q){this.A=a;this.U=b;this.da=void 0===c?null:c;this.h=void 0===d?null:d;this.P=void 0===e?null:e;this.i=void 0===f?null:f;this.j=void 0===g?null:g;this.H=void 0===h?null:h;this.N=void 0===k?null:k;this.l=void 0===l?null:l;this.m=void 0===m?null:m;this.O=void 0===q?null:q;this.R=this.C=this.G=null}Cm.prototype.size=function(){return this.U}; function Dm(a,b,c){null!=a.da&&(c.google_responsive_formats=a.da);null!=a.P&&(c.google_safe_for_responsive_override=a.P);null!=a.i&&(!0===a.i?c.google_full_width_responsive_allowed=!0:(c.google_full_width_responsive_allowed=!1,c.gfwrnwer=a.i));null!=a.j&&!0!==a.j&&(c.gfwrnher=a.j);var d=a.m||c.google_ad_width;null!=d&&(c.google_resizing_width=d);d=a.l||c.google_ad_height;null!=d&&(c.google_resizing_height=d);d=a.size().h(b);var e=a.size().height();if(!c.google_ad_resize){c.google_ad_width=d;c.google_ad_height= e;var f=a.size();b=f.h(b)+"x"+f.height();c.google_ad_format=b;c.google_responsive_auto_format=a.A;null!=a.h&&(c.armr=a.h);c.google_ad_resizable=!0;c.google_override_format=1;c.google_loader_features_used=128;!0===a.i&&(c.gfwrnh=a.size().height()+"px")}null!=a.H&&(c.gfwroml=a.H);null!=a.N&&(c.gfwromr=a.N);null!=a.l&&(c.gfwroh=a.l);null!=a.m&&(c.gfwrow=a.m);null!=a.O&&(c.gfwroz=a.O);null!=a.G&&(c.gml=a.G);null!=a.C&&(c.gmr=a.C);null!=a.R&&(c.gzi=a.R);b=Ic(window)||window;yl(b.location,"google_responsive_dummy_ad")&& (Xa([1,2,3,4,5,6,7,8],a.A)||1===a.h)&&2!==a.h&&(a=JSON.stringify({googMsgType:"adpnt",key_value:[{key:"qid",value:"DUMMY_AD"}]}),c.dash="<"+Bm+">window.top.postMessage('"+a+"', '*');\n </"+Bm+'>\n <div id="dummyAd" style="width:'+d+"px;height:"+e+'px;\n background:#ddd;border:3px solid #f00;box-sizing:border-box;\n color:#000;">\n <p>Requested size:'+d+"x"+e+"</p>\n <p>Rendered size:"+d+"x"+e+"</p>\n </div>")};var Em=["google_content_recommendation_ui_type","google_content_recommendation_columns_num","google_content_recommendation_rows_num"];function Fm(a,b){wi.call(this,a,b)}v(Fm,wi);Fm.prototype.h=function(a){return Math.min(1200,Math.max(this.minWidth(),Math.round(a)))}; function Gm(a,b){Hm(a,b);if("pedestal"==b.google_content_recommendation_ui_type)return new Cm(9,new Fm(a,Math.floor(a*b.google_phwr)));var c=Cc();468>a?c?(c=a-8-8,c=Math.floor(c/1.91+70)+Math.floor(11*(c*um.mobile_banner_image_sidebyside+wm.mobile_banner_image_sidebyside)+96),a={aa:a,$:c,J:1,K:12,T:"mobile_banner_image_sidebyside"}):(a=Am(a),a={aa:a.width,$:a.height,J:1,K:13,T:"image_sidebyside"}):(a=Am(a),a={aa:a.width,$:a.height,J:4,K:2,T:"image_stacked"});Im(b,a);return new Cm(9,new Fm(a.aa,a.$))} function Jm(a,b){Hm(a,b);var c=zm({K:b.google_content_recommendation_rows_num,J:b.google_content_recommendation_columns_num,T:b.google_content_recommendation_ui_type});if(c.M)a={aa:0,$:0,J:0,K:0,T:"image_stacked",M:c.M};else{var d=2===c.Na.length&&468<=a?1:0;var e=c.Na[d];e=0===e.indexOf("pub_control_")?e:"pub_control_"+e;var f=ym[e];for(var g=c.J[d];a/g<f&&1<g;)g--;f=g;c=c.K[d];d=Math.floor(((a-8*f-8)/f*um[e]+wm[e])*c+8*c+8);a=1500<a?{width:0,height:0,rb:"Calculated slot width is too large: "+a}: 1500<d?{width:0,height:0,rb:"Calculated slot height is too large: "+d}:{width:a,height:d};a={aa:a.width,$:a.height,J:f,K:c,T:e}}if(a.M)throw new T(a.M);Im(b,a);return new Cm(9,new Fm(a.aa,a.$))}function Hm(a,b){if(0>=a)throw new T("Invalid responsive width from Matched Content slot "+b.google_ad_slot+": "+a+". Please ensure to put this Matched Content slot into a non-zero width div container.");} function Im(a,b){a.google_content_recommendation_ui_type=b.T;a.google_content_recommendation_columns_num=b.J;a.google_content_recommendation_rows_num=b.K};function Km(a,b){wi.call(this,a,b)}v(Km,wi);Km.prototype.h=function(){return this.minWidth()};Km.prototype.i=function(a,b,c){vi(a,c);b.google_ad_resize||(c.style.height=this.height()+"px",b.rpe=!0)};var Lm={"image-top":function(a){return 600>=a?284+.414*(a-250):429},"image-middle":function(a){return 500>=a?196-.13*(a-250):164+.2*(a-500)},"image-side":function(a){return 500>=a?205-.28*(a-250):134+.21*(a-500)},"text-only":function(a){return 500>=a?187-.228*(a-250):130},"in-article":function(a){return 420>=a?a/1.2:460>=a?a/1.91+130:800>=a?a/4:200}};function Mm(a,b){wi.call(this,a,b)}v(Mm,wi);Mm.prototype.h=function(){return Math.min(1200,this.minWidth())}; function Nm(a,b,c,d,e){var f=e.google_ad_layout||"image-top";if("in-article"==f){var g=a;if("false"==e.google_full_width_responsive)a=g;else if(a=qi(b,c,g,.2,e),!0!==a)e.gfwrnwer=a,a=g;else if(a=Wf(b))if(e.google_full_width_responsive_allowed=!0,c.parentElement){b:{g=c;for(var h=0;100>h&&g.parentElement;++h){for(var k=g.parentElement.childNodes,l=0;l<k.length;++l){var m=k[l];if(m!=g&&ti(b,m))break b}g=g.parentElement;g.style.width="100%";g.style.height="auto"}}vi(b,c)}else a=g;else a=g}if(250>a)throw new T("Fluid responsive ads must be at least 250px wide: availableWidth="+ a);a=Math.min(1200,Math.floor(a));if(d&&"in-article"!=f){f=Math.ceil(d);if(50>f)throw new T("Fluid responsive ads must be at least 50px tall: height="+f);return new Cm(11,new wi(a,f))}if("in-article"!=f&&(d=e.google_ad_layout_key)){f=""+d;b=Math.pow(10,3);if(d=(c=f.match(/([+-][0-9a-z]+)/g))&&c.length){e=[];for(g=0;g<d;g++)e.push(parseInt(c[g],36)/b);b=e}else b=null;if(!b)throw new T("Invalid data-ad-layout-key value: "+f);f=(a+-725)/1E3;c=0;d=1;e=b.length;for(g=0;g<e;g++)c+=b[g]*d,d*=f;f=Math.ceil(1E3* c- -725+10);if(isNaN(f))throw new T("Invalid height: height="+f);if(50>f)throw new T("Fluid responsive ads must be at least 50px tall: height="+f);if(1200<f)throw new T("Fluid responsive ads must be at most 1200px tall: height="+f);return new Cm(11,new wi(a,f))}d=Lm[f];if(!d)throw new T("Invalid data-ad-layout value: "+f);c=Ai(c,b);b=Wf(b);b="in-article"!==f||c||a!==b?Math.ceil(d(a)):Math.ceil(1.25*d(a));return new Cm(11,"in-article"==f?new Mm(a,b):new wi(a,b))};function Om(a){return function(b){for(var c=a.length-1;0<=c;--c)if(!a[c](b))return!1;return!0}}function Pm(a,b){for(var c=Qm.slice(0),d=c.length,e=null,f=0;f<d;++f){var g=c[f];if(a(g)){if(!b||b(g))return g;null===e&&(e=g)}}return e};var W=[new V(970,90,2),new V(728,90,2),new V(468,60,2),new V(336,280,1),new V(320,100,2),new V(320,50,2),new V(300,600,4),new V(300,250,1),new V(250,250,1),new V(234,60,2),new V(200,200,1),new V(180,150,1),new V(160,600,4),new V(125,125,1),new V(120,600,4),new V(120,240,4),new V(120,120,1,!0)],Qm=[W[6],W[12],W[3],W[0],W[7],W[14],W[1],W[8],W[10],W[4],W[15],W[2],W[11],W[5],W[13],W[9],W[16]];function Rm(a,b,c,d,e){"false"==e.google_full_width_responsive?c={D:a,F:1}:"autorelaxed"==b&&e.google_full_width_responsive||Sm(b)||e.google_ad_resize?(b=ri(a,c,d,e),c=!0!==b?{D:a,F:b}:{D:Wf(c)||a,F:!0}):c={D:a,F:2};b=c.F;return!0!==b?{D:a,F:b}:d.parentElement?{D:c.D,F:b}:{D:a,F:b}} function Tm(a,b,c,d,e){var f=Wi(247,function(){return Rm(a,b,c,d,e)}),g=f.D;f=f.F;var h=!0===f,k=K(d.style.width),l=K(d.style.height),m=Um(g,b,c,d,e,h);g=m.Y;h=m.W;var q=m.pa;m=m.Ma;var t=Vm(b,q),y,F=(y=xi(d,c,"marginLeft",K))?y+"px":"",z=(y=xi(d,c,"marginRight",K))?y+"px":"";y=xi(d,c,"zIndex")||"";return new Cm(t,g,q,null,m,f,h,F,z,l,k,y)}function Sm(a){return"auto"==a||/^((^|,) *(horizontal|vertical|rectangle) *)+$/.test(a)} function Um(a,b,c,d,e,f){b="auto"==b?.25>=a/Math.min(1200,Wf(c))?4:3:pi(b);var g=!1,h=!1;if(488>Wf(c)){var k=ki(d,c);var l=Ai(d,c);g=!l&&k;h=l&&k}l=[yi(a),sm(b)];l.push(zi(488>Wf(c),c,d,h));null!=e.google_max_responsive_height&&l.push(Ci(e.google_max_responsive_height));var m=[function(t){return!t.ib}];if(g||h)g=Di(c,d),m.push(Ci(g));var q=Pm(Om(l),Om(m));if(!q)throw new T("No slot size for availableWidth="+a);l=Wi(248,function(){var t;a:if(f){if(e.gfwrnh&&(t=K(e.gfwrnh))){t={Y:new Km(a,t),W:!0}; break a}t=a/1.2;var y=Math;var F=y.min;if(e.google_resizing_allowed||"true"==e.google_full_width_responsive)var z=Infinity;else{z=d;var E=Infinity;do{var S=xi(z,c,"height",K);S&&(E=Math.min(E,S));(S=xi(z,c,"maxHeight",K))&&(E=Math.min(E,S))}while((z=z.parentElement)&&"HTML"!=z.tagName);z=E}y=F.call(y,t,z);if(y<.5*t||100>y)y=t;P(hf)&&!Ai(d,c)&&(y=Math.max(y,.5*Vf(c).clientHeight));t={Y:new Km(a,Math.floor(y)),W:y<t?102:!0}}else t={Y:q,W:100};return t});g=l.Y;l=l.W;return"in-article"===e.google_ad_layout&& Wm(c)?{Y:Xm(a,c,d,g,e),W:!1,pa:b,Ma:k}:{Y:g,W:l,pa:b,Ma:k}}function Vm(a,b){if("auto"==a)return 1;switch(b){case 2:return 2;case 1:return 3;case 4:return 4;case 3:return 5;case 6:return 6;case 5:return 7;case 7:return 8}throw Error("bad mask");}function Xm(a,b,c,d,e){var f=e.google_ad_height||xi(c,b,"height",K);b=Nm(a,b,c,f,e).size();return b.minWidth()*b.height()>a*d.height()?new V(b.minWidth(),b.height(),1):d}function Wm(a){return P(ff)||a.location&&"#hffwroe2etoq"==a.location.hash};function Ym(a,b,c,d,e){var f;(f=Wf(b))?488>Wf(b)?b.innerHeight>=b.innerWidth?(e.google_full_width_responsive_allowed=!0,vi(b,c),f={D:f,F:!0}):f={D:a,F:5}:f={D:a,F:4}:f={D:a,F:10};var g=f;f=g.D;g=g.F;if(!0!==g||a==f)return new Cm(12,new wi(a,d),null,null,!0,g,100);a=Um(f,"auto",b,c,e,!0);return new Cm(1,a.Y,a.pa,2,!0,g,a.W)};function Zm(a,b){var c=b.google_ad_format;if("autorelaxed"==c){a:{if("pedestal"!=b.google_content_recommendation_ui_type)for(a=u(Em),c=a.next();!c.done;c=a.next())if(null!=b[c.value]){b=!0;break a}b=!1}return b?9:5}if(Sm(c))return 1;if("link"===c)return 4;if("fluid"==c){if(c="in-article"===b.google_ad_layout)c=P(gf)||P(ff)||a.location&&("#hffwroe2etop"==a.location.hash||"#hffwroe2etoq"==a.location.hash);return c?($m(b),1):8}if(27===b.google_reactive_ad_format)return $m(b),1} function an(a,b,c,d,e){e=b.offsetWidth||(c.google_ad_resize||(void 0===e?!1:e))&&xi(b,d,"width",K)||c.google_ad_width||0;4===a&&(c.google_ad_format="auto",a=1);var f=(f=bn(a,e,b,c,d))?f:Tm(e,c.google_ad_format,d,b,c);f.size().i(d,c,b);Dm(f,e,c);1!=a&&(a=f.size().height(),b.style.height=a+"px")} function bn(a,b,c,d,e){var f=d.google_ad_height||xi(c,e,"height",K);switch(a){case 5:return f=Wi(247,function(){return Rm(b,d.google_ad_format,e,c,d)}),a=f.D,f=f.F,!0===f&&b!=a&&vi(e,c),!0===f?d.google_full_width_responsive_allowed=!0:(d.google_full_width_responsive_allowed=!1,d.gfwrnwer=f),Gm(a,d);case 9:return Jm(b,d);case 8:return Nm(b,e,c,f,d);case 10:return Ym(b,e,c,f,d)}}function $m(a){a.google_ad_format="auto";a.armr=3};function cn(a,b){var c=Ic(b);if(c){c=Wf(c);var d=Nc(a,b)||{},e=d.direction;if("0px"===d.width&&"none"!==d.cssFloat)return-1;if("ltr"===e&&c)return Math.floor(Math.min(1200,c-a.getBoundingClientRect().left));if("rtl"===e&&c)return a=b.document.body.getBoundingClientRect().right-a.getBoundingClientRect().right,Math.floor(Math.min(1200,c-a-Math.floor((c-b.document.body.clientWidth)/2)))}return-1};var dn=ja(["https://pagead2.googlesyndication.com/pagead/managed/js/adsense/","/slotcar_library",".js"]),en=ja(["https://googleads.g.doubleclick.net/pagead/html/","/","/zrt_lookup.html"]),fn=ja(["https://pagead2.googlesyndication.com/pagead/managed/js/adsense/","/show_ads_impl",".js"]),gn=ja(["https://pagead2.googlesyndication.com/pagead/managed/js/adsense/","/show_ads_impl_with_ama",".js"]),hn=ja(["https://pagead2.googlesyndication.com/pagead/managed/js/adsense/","/show_ads_impl_instrumented",".js"]);function jn(a){Ui.Ta(function(b){b.shv=String(a);b.mjsv="m202204040101";var c=O(Oh).h(),d=U(w);d.eids||(d.eids=[]);b.eid=c.concat(d.eids).join(",")})};function kn(a){var b=a.nb;return a.eb||("dev"===b?"dev":"")};var ln={},mn=(ln.google_ad_modifications=!0,ln.google_analytics_domain_name=!0,ln.google_analytics_uacct=!0,ln.google_pause_ad_requests=!0,ln.google_user_agent_client_hint=!0,ln);function nn(a){return(a=a.innerText||a.innerHTML)&&(a=a.replace(/^\s+/,"").split(/\r?\n/,1)[0].match(/^\x3c!--+(.*?)(?:--+>)?\s*$/))&&RegExp("google_ad_client").test(a[1])?a[1]:null} function on(a){if(a=a.innerText||a.innerHTML)if(a=a.replace(/^\s+|\s+$/g,"").replace(/\s*(\r?\n)+\s*/g,";"),(a=a.match(/^\x3c!--+(.*?)(?:--+>)?$/)||a.match(/^\/*\s*<!\[CDATA\[(.*?)(?:\/*\s*\]\]>)?$/i))&&RegExp("google_ad_client").test(a[1]))return a[1];return null} function pn(a){switch(a){case "true":return!0;case "false":return!1;case "null":return null;case "undefined":break;default:try{var b=a.match(/^(?:'(.*)'|"(.*)")$/);if(b)return b[1]||b[2]||"";if(/^[-+]?\d*(\.\d+)?$/.test(a)){var c=parseFloat(a);return c===c?c:void 0}}catch(d){}}};function qn(a){if(a.google_ad_client)return String(a.google_ad_client);var b,c,d,e,f;if(null!=(e=null!=(d=null==(b=U(a).head_tag_slot_vars)?void 0:b.google_ad_client)?d:null==(c=a.document.querySelector(".adsbygoogle[data-ad-client]"))?void 0:c.getAttribute("data-ad-client")))b=e;else{b:{b=a.document.getElementsByTagName("script");a=a.navigator&&a.navigator.userAgent||"";a=RegExp("appbankapppuzdradb|daumapps|fban|fbios|fbav|fb_iab|gsa/|messengerforios|naver|niftyappmobile|nonavigation|pinterest|twitter|ucbrowser|yjnewsapp|youtube", "i").test(a)||/i(phone|pad|pod)/i.test(a)&&/applewebkit/i.test(a)&&!/version|safari/i.test(a)&&!qd()?nn:on;for(c=b.length-1;0<=c;c--)if(d=b[c],!d.google_parsed_script_for_pub_code&&(d.google_parsed_script_for_pub_code=!0,d=a(d))){b=d;break b}b=null}if(b){a=/(google_\w+) *= *(['"]?[\w.-]+['"]?) *(?:;|$)/gm;for(c={};d=a.exec(b);)c[d[1]]=pn(d[2]);b=c.google_ad_client?c.google_ad_client:""}else b=""}return null!=(f=b)?f:""};var rn="undefined"===typeof sttc?void 0:sttc;function sn(a){var b=Ui;try{return Yf(a,Zf),new Rk(JSON.parse(a))}catch(c){b.I(838,c instanceof Error?c:Error(String(c)),void 0,function(d){d.jspb=String(a)})}return new Rk};var tn=O(yf).h(mf.h,mf.defaultValue);function un(){var a=L.document;a=void 0===a?window.document:a;ed(tn,a)};var vn=O(yf).h(nf.h,nf.defaultValue);function wn(){var a=L.document;a=void 0===a?window.document:a;ed(vn,a)};var xn=ja(["https://pagead2.googlesyndication.com/pagead/js/err_rep.js"]);function yn(){this.h=null;this.j=!1;this.l=Math.random();this.i=this.I;this.m=null}n=yn.prototype;n.Ta=function(a){this.h=a};n.Va=function(a){this.j=a};n.Ua=function(a){this.i=a}; n.I=function(a,b,c,d,e){if((this.j?this.l:Math.random())>(void 0===c?.01:c))return!1;Rh(b)||(b=new Qh(b,{context:a,id:void 0===e?"jserror":e}));if(d||this.h)b.meta={},this.h&&this.h(b.meta),d&&d(b.meta);w.google_js_errors=w.google_js_errors||[];w.google_js_errors.push(b);if(!w.error_rep_loaded){a=nd(xn);var f;Lc(w.document,null!=(f=this.m)?f:hc(qc(a).toString()));w.error_rep_loaded=!0}return!1};n.oa=function(a,b,c){try{return b()}catch(d){if(!this.i(a,d,.01,c,"jserror"))throw d;}}; n.Oa=function(a,b){var c=this;return function(){var d=ta.apply(0,arguments);return c.oa(a,function(){return b.apply(void 0,d)})}};n.Pa=function(a,b){var c=this;b.catch(function(d){d=d?d:"unknown rejection";c.I(a,d instanceof Error?d:Error(d))})};function zn(a,b,c){var d=window;return function(){var e=Yh(),f=3;try{var g=b.apply(this,arguments)}catch(h){f=13;if(c)return c(a,h),g;throw h;}finally{d.google_measure_js_timing&&e&&(e={label:a.toString(),value:e,duration:(Yh()||0)-e,type:f},f=d.google_js_reporting_queue=d.google_js_reporting_queue||[],2048>f.length&&f.push(e))}return g}}function An(a,b){return zn(a,b,function(c,d){(new yn).I(c,d)})};function Bn(a,b){return null==b?"&"+a+"=null":"&"+a+"="+Math.floor(b)}function Cn(a,b){return"&"+a+"="+b.toFixed(3)}function Dn(){var a=new p.Set,b=aj();try{if(!b)return a;for(var c=b.pubads(),d=u(c.getSlots()),e=d.next();!e.done;e=d.next())a.add(e.value.getSlotId().getDomId())}catch(f){}return a}function En(a){a=a.id;return null!=a&&(Dn().has(a)||r(a,"startsWith").call(a,"google_ads_iframe_")||r(a,"startsWith").call(a,"aswift"))} function Fn(a,b,c){if(!a.sources)return!1;switch(Gn(a)){case 2:var d=Hn(a);if(d)return c.some(function(f){return In(d,f)});case 1:var e=Jn(a);if(e)return b.some(function(f){return In(e,f)})}return!1}function Gn(a){if(!a.sources)return 0;a=a.sources.filter(function(b){return b.previousRect&&b.currentRect});if(1<=a.length){a=a[0];if(a.previousRect.top<a.currentRect.top)return 2;if(a.previousRect.top>a.currentRect.top)return 1}return 0}function Jn(a){return Kn(a,function(b){return b.currentRect})} function Hn(a){return Kn(a,function(b){return b.previousRect})}function Kn(a,b){return a.sources.reduce(function(c,d){d=b(d);return c?d&&0!==d.width*d.height?d.top<c.top?d:c:c:d},null)} function Ln(){Mj.call(this);this.i=this.h=this.P=this.O=this.H=0;this.Ba=this.ya=Number.NEGATIVE_INFINITY;this.ua=this.wa=this.xa=this.za=this.Ea=this.m=this.Da=this.U=0;this.va=!1;this.R=this.N=this.C=0;var a=document.querySelector("[data-google-query-id]");this.Ca=a?a.getAttribute("data-google-query-id"):null;this.l=null;this.Aa=!1;this.ga=function(){}}v(Ln,Mj); function Mn(){var a=new Ln;if(P(of)){var b=window;if(!b.google_plmetrics&&window.PerformanceObserver){b.google_plmetrics=!0;b=u(["layout-shift","largest-contentful-paint","first-input","longtask"]);for(var c=b.next();!c.done;c=b.next())c=c.value,Nn(a).observe({type:c,buffered:!0});On(a)}}} function Nn(a){a.l||(a.l=new PerformanceObserver(An(640,function(b){var c=Pn!==window.scrollX||Qn!==window.scrollY?[]:Rn,d=Sn();b=u(b.getEntries());for(var e=b.next();!e.done;e=b.next())switch(e=e.value,e.entryType){case "layout-shift":var f=a;if(!e.hadRecentInput){f.H+=Number(e.value);Number(e.value)>f.O&&(f.O=Number(e.value));f.P+=1;var g=Fn(e,c,d);g&&(f.m+=e.value,f.za++);if(5E3<e.startTime-f.ya||1E3<e.startTime-f.Ba)f.ya=e.startTime,f.h=0,f.i=0;f.Ba=e.startTime;f.h+=e.value;g&&(f.i+=e.value); f.h>f.U&&(f.U=f.h,f.Ea=f.i,f.Da=e.startTime+e.duration)}break;case "largest-contentful-paint":a.xa=Math.floor(e.renderTime||e.loadTime);a.wa=e.size;break;case "first-input":a.ua=Number((e.processingStart-e.startTime).toFixed(3));a.va=!0;break;case "longtask":e=Math.max(0,e.duration-50),a.C+=e,a.N=Math.max(a.N,e),a.R+=1}})));return a.l} function On(a){var b=An(641,function(){var d=document;2==(d.prerendering?3:{visible:1,hidden:2,prerender:3,preview:4,unloaded:5}[d.visibilityState||d.webkitVisibilityState||d.mozVisibilityState||""]||0)&&Tn(a)}),c=An(641,function(){return void Tn(a)});document.addEventListener("visibilitychange",b);document.addEventListener("unload",c);a.ga=function(){document.removeEventListener("visibilitychange",b);document.removeEventListener("unload",c);Nn(a).disconnect()}} Ln.prototype.j=function(){Mj.prototype.j.call(this);this.ga()}; function Tn(a){if(!a.Aa){a.Aa=!0;Nn(a).takeRecords();var b="https://pagead2.googlesyndication.com/pagead/gen_204?id=plmetrics";window.LayoutShift&&(b+=Cn("cls",a.H),b+=Cn("mls",a.O),b+=Bn("nls",a.P),window.LayoutShiftAttribution&&(b+=Cn("cas",a.m),b+=Bn("nas",a.za)),b+=Cn("wls",a.U),b+=Cn("tls",a.Da),window.LayoutShiftAttribution&&(b+=Cn("was",a.Ea)));window.LargestContentfulPaint&&(b+=Bn("lcp",a.xa),b+=Bn("lcps",a.wa));window.PerformanceEventTiming&&a.va&&(b+=Bn("fid",a.ua));window.PerformanceLongTaskTiming&& (b+=Bn("cbt",a.C),b+=Bn("mbt",a.N),b+=Bn("nlt",a.R));for(var c=0,d=u(document.getElementsByTagName("iframe")),e=d.next();!e.done;e=d.next())En(e.value)&&c++;b+=Bn("nif",c);b+=Bn("ifi",pd(window));c=O(Oh).h();b+="&eid="+encodeURIComponent(c.join());b+="&top="+(w===w.top?1:0);b+=a.Ca?"&qqid="+encodeURIComponent(a.Ca):Bn("pvsid",fd(w));window.googletag&&(b+="&gpt=1");window.fetch(b,{keepalive:!0,credentials:"include",redirect:"follow",method:"get",mode:"no-cors"});a.A||(a.A=!0,a.j())}} function In(a,b){var c=Math.min(a.right,b.right)-Math.max(a.left,b.left);a=Math.min(a.bottom,b.bottom)-Math.max(a.top,b.top);return 0>=c||0>=a?!1:50<=100*c*a/((b.right-b.left)*(b.bottom-b.top))} function Sn(){var a=[].concat(ka(document.getElementsByTagName("iframe"))).filter(En),b=[].concat(ka(Dn())).map(function(c){return document.getElementById(c)}).filter(function(c){return null!==c});Pn=window.scrollX;Qn=window.scrollY;return Rn=[].concat(ka(a),ka(b)).map(function(c){return c.getBoundingClientRect()})}var Pn=void 0,Qn=void 0,Rn=[];var X={issuerOrigin:"https://attestation.android.com",issuancePath:"/att/i",redemptionPath:"/att/r"},Y={issuerOrigin:"https://pagead2.googlesyndication.com",issuancePath:"/dtt/i",redemptionPath:"/dtt/r",getStatePath:"/dtt/s"};var Un=O(yf).h(wf.h,wf.defaultValue); function Vn(a,b,c){Mj.call(this);var d=this;this.i=a;this.h=[];b&&Wn()&&this.h.push(X);c&&this.h.push(Y);if(document.hasTrustToken&&!P(tf)){var e=new p.Map;this.h.forEach(function(f){e.set(f.issuerOrigin,{issuerOrigin:f.issuerOrigin,state:d.i?1:12,hasRedemptionRecord:!1})});window.goog_tt_state_map=window.goog_tt_state_map&&window.goog_tt_state_map instanceof p.Map?new p.Map([].concat(ka(e),ka(window.goog_tt_state_map))):e;window.goog_tt_promise_map&&window.goog_tt_promise_map instanceof p.Map||(window.goog_tt_promise_map= new p.Map)}}v(Vn,Mj);function Wn(){var a=void 0===a?window:a;a=a.navigator.userAgent;var b=/Chrome/.test(a);return/Android/.test(a)&&b}function Xn(){var a=void 0===a?window.document:a;ed(Un,a)}function Yn(a,b){return a||".google.ch"===b||"function"===typeof L.__tcfapi}function Z(a,b,c){var d,e=null==(d=window.goog_tt_state_map)?void 0:d.get(a);e&&(e.state=b,void 0!=c&&(e.hasRedemptionRecord=c))} function Zn(){var a=X.issuerOrigin+X.redemptionPath,b={keepalive:!0,trustToken:{type:"token-redemption",issuer:X.issuerOrigin,refreshPolicy:"none"}};Z(X.issuerOrigin,2);return window.fetch(a,b).then(function(c){if(!c.ok)throw Error(c.status+": Network response was not ok!");Z(X.issuerOrigin,6,!0)}).catch(function(c){c&&"NoModificationAllowedError"===c.name?Z(X.issuerOrigin,6,!0):Z(X.issuerOrigin,5)})} function $n(){var a=X.issuerOrigin+X.issuancePath;Z(X.issuerOrigin,8);return window.fetch(a,{keepalive:!0,trustToken:{type:"token-request"}}).then(function(b){if(!b.ok)throw Error(b.status+": Network response was not ok!");Z(X.issuerOrigin,10);return Zn()}).catch(function(b){if(b&&"NoModificationAllowedError"===b.name)return Z(X.issuerOrigin,10),Zn();Z(X.issuerOrigin,9)})}function ao(){Z(X.issuerOrigin,13);return document.hasTrustToken(X.issuerOrigin).then(function(a){return a?Zn():$n()})} function bo(){Z(Y.issuerOrigin,13);if(p.Promise){var a=document.hasTrustToken(Y.issuerOrigin).then(function(e){return e}).catch(function(e){return p.Promise.reject({state:19,error:e})}),b=Y.issuerOrigin+Y.redemptionPath,c={keepalive:!0,trustToken:{type:"token-redemption",refreshPolicy:"none"}};Z(Y.issuerOrigin,16);a=a.then(function(e){return window.fetch(b,c).then(function(f){if(!f.ok)throw Error(f.status+": Network response was not ok!");Z(Y.issuerOrigin,18,!0)}).catch(function(f){if(f&&"NoModificationAllowedError"=== f.name)Z(Y.issuerOrigin,18,!0);else{if(e)return p.Promise.reject({state:17,error:f});Z(Y.issuerOrigin,17)}})}).then(function(){return document.hasTrustToken(Y.issuerOrigin).then(function(e){return e}).catch(function(e){return p.Promise.reject({state:19,error:e})})}).then(function(e){var f=Y.issuerOrigin+Y.getStatePath;Z(Y.issuerOrigin,20);return window.fetch(f+"?ht="+e,{trustToken:{type:"send-redemption-record",issuers:[Y.issuerOrigin]}}).then(function(g){if(!g.ok)throw Error(g.status+": Network response was not ok!"); Z(Y.issuerOrigin,22);return g.text().then(function(h){return JSON.parse(h)})}).catch(function(g){return p.Promise.reject({state:21,error:g})})});var d=fd(window);return a.then(function(e){var f=Y.issuerOrigin+Y.issuancePath;return e&&e.srqt&&e.cs?(Z(Y.issuerOrigin,23),window.fetch(f+"?cs="+e.cs+"&correlator="+d,{keepalive:!0,trustToken:{type:"token-request"}}).then(function(g){if(!g.ok)throw Error(g.status+": Network response was not ok!");Z(Y.issuerOrigin,25);return e}).catch(function(g){return p.Promise.reject({state:24, error:g})})):e}).then(function(e){if(e&&e.srdt&&e.cs)return Z(Y.issuerOrigin,26),window.fetch(b+"?cs="+e.cs+"&correlator="+d,{keepalive:!0,trustToken:{type:"token-redemption",refreshPolicy:"refresh"}}).then(function(f){if(!f.ok)throw Error(f.status+": Network response was not ok!");Z(Y.issuerOrigin,28,!0)}).catch(function(f){return p.Promise.reject({state:27,error:f})})}).then(function(){Z(Y.issuerOrigin,29)}).catch(function(e){if(e instanceof Object&&e.hasOwnProperty("state")&&e.hasOwnProperty("error"))if("number"=== typeof e.state&&e.error instanceof Error){Z(Y.issuerOrigin,e.state);var f=Q(vf);Math.random()<=f&&Ff({state:e.state,err:e.error.toString()})}else throw Error(e);else throw e;})}} function co(a){if(document.hasTrustToken&&!P(tf)&&a.i){var b=window.goog_tt_promise_map;if(b&&b instanceof p.Map){var c=[];if(a.h.some(function(e){return e.issuerOrigin===X.issuerOrigin})){var d=b.get(X.issuerOrigin);d||(d=ao(),b.set(X.issuerOrigin,d));c.push(d)}a.h.some(function(e){return e.issuerOrigin===Y.issuerOrigin})&&(a=b.get(Y.issuerOrigin),a||(a=bo(),b.set(Y.issuerOrigin,a)),c.push(a));if(0<c.length&&p.Promise&&p.Promise.all)return p.Promise.all(c)}}};function eo(a){J.call(this,a,-1,fo)}v(eo,J);function go(a,b){return B(a,2,b)}function ho(a,b){return B(a,3,b)}function io(a,b){return B(a,4,b)}function jo(a,b){return B(a,5,b)}function ko(a,b){return B(a,9,b)}function lo(a,b){return Gb(a,10,b)}function mo(a,b){return B(a,11,b)}function no(a,b){return B(a,1,b)}function oo(a){J.call(this,a)}v(oo,J);oo.prototype.getVersion=function(){return I(this,2)};var fo=[10,6];var po="platform platformVersion architecture model uaFullVersion bitness fullVersionList wow64".split(" ");function qo(){var a;return null!=(a=L.google_tag_data)?a:L.google_tag_data={}} function ro(){var a,b;if("function"!==typeof(null==(a=L.navigator)?void 0:null==(b=a.userAgentData)?void 0:b.getHighEntropyValues))return null;var c=qo();if(c.uach_promise)return c.uach_promise;a=L.navigator.userAgentData.getHighEntropyValues(po).then(function(d){null!=c.uach||(c.uach=d);return d});return c.uach_promise=a} function so(a){var b;return mo(lo(ko(jo(io(ho(go(no(new eo,a.platform||""),a.platformVersion||""),a.architecture||""),a.model||""),a.uaFullVersion||""),a.bitness||""),(null==(b=a.fullVersionList)?void 0:b.map(function(c){var d=new oo;d=B(d,1,c.brand);return B(d,2,c.version)}))||[]),a.wow64||!1)} function to(){if(P(pf)){var a,b;return null!=(b=null==(a=ro())?void 0:a.then(function(f){return so(f)}))?b:null}var c,d;if("function"!==typeof(null==(c=L.navigator)?void 0:null==(d=c.userAgentData)?void 0:d.getHighEntropyValues))return null;var e;return null!=(e=L.navigator.userAgentData.getHighEntropyValues(po).then(function(f){return so(f)}))?e:null};function uo(a,b){b.google_ad_host||(a=vo(a))&&(b.google_ad_host=a)}function wo(a,b,c){c=void 0===c?"":c;L.google_sa_impl&&!L.document.getElementById("google_shimpl")&&(delete L.google_sa_queue,delete L.google_sa_impl);L.google_sa_queue||(L.google_sa_queue=[],L.google_process_slots=Xi(215,function(){return xo(L.google_sa_queue)}),a=yo(c,a,b),Lc(L.document,a).id="google_shimpl")} function xo(a){var b=a.shift();"function"===typeof b&&Wi(216,b);a.length&&w.setTimeout(Xi(215,function(){return xo(a)}),0)}function zo(a,b,c){a.google_sa_queue=a.google_sa_queue||[];a.google_sa_impl?c(b):a.google_sa_queue.push(b)} function yo(a,b,c){var d=Math.random()<Q(bf)?hc(qc(b.pb).toString()):null;b=D(c,4)?b.ob:b.qb;d=d?d:hc(qc(b).toString());b={};a:{if(D(c,4)){if(c=a||qn(L)){var e={};c=(e.client=c,e.plah=L.location.host,e);break a}throw Error("PublisherCodeNotFoundForAma");}c={}}Ao(c,b);a:{if(P($e)||P(Oe)){a=a||qn(L);var f;var g=(c=null==(g=U(L))?void 0:null==(f=g.head_tag_slot_vars)?void 0:f.google_ad_host)?c:vo(L);if(a){f={};g=(f.client=a,f.plah=L.location.host,f.ama_t="adsense",f.asntp=Q(Ge),f.asntpv=Q(Ke),f.asntpl= Q(Ie),f.asntpm=Q(Je),f.asntpc=Q(He),f.asna=Q(Ce),f.asnd=Q(De),f.asnp=Q(Ee),f.asns=Q(Fe),f.asmat=Q(Be),f.asptt=Q(Le),f.easpi=P($e),f.asro=P(Me),f.host=g,f.easai=P(Ze),f);break a}}g={}}Ao(g,b);Ao(zf()?{bust:zf()}:{},b);return ec(d,b)}function Ao(a,b){Sc(a,function(c,d){void 0===b[d]&&(b[d]=c)})}function vo(a){if(a=a.document.querySelector('meta[name="google-adsense-platform-account"]'))return a.getAttribute("content")} function Bo(a){a:{var b=void 0===b?!1:b;var c=void 0===c?1024:c;for(var d=[w.top],e=[],f=0,g;g=d[f++];){b&&!Hc(g)||e.push(g);try{if(g.frames)for(var h=0;h<g.frames.length&&d.length<c;++h)d.push(g.frames[h])}catch(l){}}for(b=0;b<e.length;b++)try{var k=e[b].frames.google_esf;if(k){id=k;break a}}catch(l){}id=null}if(id)return null;e=Mc("IFRAME");e.id="google_esf";e.name="google_esf";e.src=sc(a.vb);e.style.display="none";return e} function Co(a,b,c,d){Do(a,b,c,d,function(e,f){e=e.document;for(var g=void 0,h=0;!g||e.getElementById(g+"_anchor");)g="aswift_"+h++;e=g;g=Number(f.google_ad_width||0);f=Number(f.google_ad_height||0);h=Mc("INS");h.id=e+"_anchor";pm(h,g,f);h.style.display="block";var k=Mc("INS");k.id=e+"_expand";pm(k,g,f);k.style.display="inline-table";k.appendChild(h);c.appendChild(k);return e})} function Do(a,b,c,d,e){e=e(a,b);Eo(a,c,b);c=Ia;var f=(new Date).getTime();b.google_lrv=I(d,2);b.google_async_iframe_id=e;b.google_start_time=c;b.google_bpp=f>c?f-c:1;a.google_sv_map=a.google_sv_map||{};a.google_sv_map[e]=b;d=a.document.getElementById(e+"_anchor")?function(h){return h()}:function(h){return window.setTimeout(h,0)};var g={pubWin:a,vars:b};zo(a,function(){var h=a.google_sa_impl(g);h&&h.catch&&Zi(911,h)},d)} function Eo(a,b,c){var d=c.google_ad_output,e=c.google_ad_format,f=c.google_ad_width||0,g=c.google_ad_height||0;e||"html"!=d&&null!=d||(e=f+"x"+g);d=!c.google_ad_slot||c.google_override_format||!qm[c.google_ad_width+"x"+c.google_ad_height]&&"aa"==c.google_loader_used;e&&d?e=e.toLowerCase():e="";c.google_ad_format=e;if("number"!==typeof c.google_reactive_sra_index||!c.google_ad_unit_key){e=[c.google_ad_slot,c.google_orig_ad_format||c.google_ad_format,c.google_ad_type,c.google_orig_ad_width||c.google_ad_width, c.google_orig_ad_height||c.google_ad_height];d=[];f=0;for(g=b;g&&25>f;g=g.parentNode,++f)9===g.nodeType?d.push(""):d.push(g.id);(d=d.join())&&e.push(d);c.google_ad_unit_key=Tc(e.join(":")).toString();var h=void 0===h?!1:h;e=[];for(d=0;b&&25>d;++d){f="";void 0!==h&&h||(f=(f=9!==b.nodeType&&b.id)?"/"+f:"");a:{if(b&&b.nodeName&&b.parentElement){g=b.nodeName.toString().toLowerCase();for(var k=b.parentElement.childNodes,l=0,m=0;m<k.length;++m){var q=k[m];if(q.nodeName&&q.nodeName.toString().toLowerCase()=== g){if(b===q){g="."+l;break a}++l}}}g=""}e.push((b.nodeName&&b.nodeName.toString().toLowerCase())+f+g);b=b.parentElement}h=e.join()+":";b=[];if(a)try{var t=a.parent;for(e=0;t&&t!==a&&25>e;++e){var y=t.frames;for(d=0;d<y.length;++d)if(a===y[d]){b.push(d);break}a=t;t=a.parent}}catch(F){}c.google_ad_dom_fingerprint=Tc(h+b.join()).toString()}}function Fo(){var a=Ic(w);a&&(a=Uf(a),a.tagSpecificState[1]||(a.tagSpecificState[1]={debugCard:null,debugCardRequested:!1}))} function Go(a){Xn();Yn(Wk(),I(a,8))||Xi(779,function(){var b=window;b=void 0===b?window:b;b=P(b.PeriodicSyncManager?rf:sf);var c=P(uf);b=new Vn(!0,b,c);0<Q(xf)?L.google_trust_token_operation_promise=co(b):co(b)})();a=to();null!=a&&a.then(function(b){L.google_user_agent_client_hint=Lb(b)});wn();un()};function Ho(a,b){switch(a){case "google_reactive_ad_format":return a=parseInt(b,10),isNaN(a)?0:a;case "google_allow_expandable_ads":return/^true$/.test(b);default:return b}} function Io(a,b){if(a.getAttribute("src")){var c=a.getAttribute("src")||"";(c=Gc(c))&&(b.google_ad_client=Ho("google_ad_client",c))}a=a.attributes;c=a.length;for(var d=0;d<c;d++){var e=a[d];if(/data-/.test(e.name)){var f=Ja(e.name.replace("data-matched-content","google_content_recommendation").replace("data","google").replace(/-/g,"_"));b.hasOwnProperty(f)||(e=Ho(f,e.value),null!==e&&(b[f]=e))}}} function Jo(a){if(a=ld(a))switch(a.data&&a.data.autoFormat){case "rspv":return 13;case "mcrspv":return 15;default:return 14}else return 12} function Ko(a,b,c,d){Io(a,b);if(c.document&&c.document.body&&!Zm(c,b)&&!b.google_reactive_ad_format){var e=parseInt(a.style.width,10),f=cn(a,c);if(0<f&&e>f){var g=parseInt(a.style.height,10);e=!!qm[e+"x"+g];var h=f;if(e){var k=rm(f,g);if(k)h=k,b.google_ad_format=k+"x"+g+"_0ads_al";else throw new T("No slot size for availableWidth="+f);}b.google_ad_resize=!0;b.google_ad_width=h;e||(b.google_ad_format=null,b.google_override_format=!0);f=h;a.style.width=f+"px";g=Tm(f,"auto",c,a,b);h=f;g.size().i(c,b, a);Dm(g,h,b);g=g.size();b.google_responsive_formats=null;g.minWidth()>f&&!e&&(b.google_ad_width=g.minWidth(),a.style.width=g.minWidth()+"px")}}e=a.offsetWidth||xi(a,c,"width",K)||b.google_ad_width||0;f=Fa(Tm,e,"auto",c,a,b,!1,!0);if(!P(Xe)&&488>Wf(c)){g=Ic(c)||c;h=b.google_ad_client;d=g.location&&"#ftptohbh"===g.location.hash?2:yl(g.location,"google_responsive_slot_preview")||P(ef)?1:P(df)?2:Yk(g,1,h,d)?1:0;if(g=0!==d)b:if(b.google_reactive_ad_format||Zm(c,b)||mi(a,b))g=!1;else{for(g=a;g;g=g.parentElement){h= Nc(g,c);if(!h){b.gfwrnwer=18;g=!1;break b}if(!Xa(["static","relative"],h.position)){b.gfwrnwer=17;g=!1;break b}}g=qi(c,a,e,.3,b);!0!==g?(b.gfwrnwer=g,g=!1):g=c===c.top?!0:!1}g?(b.google_resizing_allowed=!0,b.ovlp=!0,2===d?(d={},Dm(f(),e,d),b.google_resizing_width=d.google_ad_width,b.google_resizing_height=d.google_ad_height,b.iaaso=!1):(b.google_ad_format="auto",b.iaaso=!0,b.armr=1),d=!0):d=!1}else d=!1;if(e=Zm(c,b))an(e,a,b,c,d);else{if(mi(a,b)){if(d=Nc(a,c))a.style.width=d.width,a.style.height= d.height,li(d,b);b.google_ad_width||(b.google_ad_width=a.offsetWidth);b.google_ad_height||(b.google_ad_height=a.offsetHeight);b.google_loader_features_used=256;b.google_responsive_auto_format=Jo(c)}else li(a.style,b);c.location&&"#gfwmrp"==c.location.hash||12==b.google_responsive_auto_format&&"true"==b.google_full_width_responsive?an(10,a,b,c,!1):.01>Math.random()&&12===b.google_responsive_auto_format&&(a=ri(a.offsetWidth||parseInt(a.style.width,10)||b.google_ad_width,c,a,b),!0!==a?(b.efwr=!1,b.gfwrnwer= a):b.efwr=!0)}};function Lo(a){this.j=new p.Set;this.u=md()||window;this.h=Q(ze);var b=0<this.h&&Rc()<1/this.h;this.A=(this.i=!!Hj(Dj(),30,b))?fd(this.u):0;this.m=this.i?qn(this.u):"";this.l=null!=a?a:new yg(100)}function Mo(){var a=O(Lo);var b=new qk;b=B(b,1,Vf(a.u).scrollWidth);b=B(b,2,Vf(a.u).scrollHeight);var c=new qk;c=B(c,1,Wf(a.u));c=B(c,2,Vf(a.u).clientHeight);var d=new sk;d=B(d,1,a.A);d=B(d,2,a.m);d=B(d,3,a.h);var e=new rk;b=Eb(e,2,b);b=Eb(b,1,c);b=Fb(d,4,tk,b);a.i&&!a.j.has(1)&&(a.j.add(1),ug(a.l,b))};function No(a){var b=window;var c=void 0===c?null:c;xc(b,"message",function(d){try{var e=JSON.parse(d.data)}catch(f){return}!e||"sc-cnf"!==e.googMsgType||c&&/[:|%3A]javascript\(/i.test(d.data)&&!c(e,d)||a(e,d)})};function Oo(a,b){b=void 0===b?500:b;Mj.call(this);this.i=a;this.ta=b;this.h=null;this.m={};this.l=null}v(Oo,Mj);Oo.prototype.j=function(){this.m={};this.l&&(yc(this.i,this.l),delete this.l);delete this.m;delete this.i;delete this.h;Mj.prototype.j.call(this)};function Po(a){Mj.call(this);this.h=a;this.i=null;this.l=!1}v(Po,Mj);var Qo=null,Ro=[],So=new p.Map,To=-1;function Uo(a){return Fi.test(a.className)&&"done"!=a.dataset.adsbygoogleStatus}function Vo(a,b,c){a.dataset.adsbygoogleStatus="done";Wo(a,b,c)} function Wo(a,b,c){var d=window;d.google_spfd||(d.google_spfd=Ko);var e=b.google_reactive_ads_config;e||Ko(a,b,d,c);uo(d,b);if(!Xo(a,b,d)){e||(d.google_lpabyc=ni(a,d)+xi(a,d,"height",K));if(e){e=e.page_level_pubvars||{};if(U(L).page_contains_reactive_tag&&!U(L).allow_second_reactive_tag){if(e.pltais){wl(!1);return}throw new T("Only one 'enable_page_level_ads' allowed per page.");}U(L).page_contains_reactive_tag=!0;wl(7===e.google_pgb_reactive)}b.google_unique_id=od(d);Sc(mn,function(f,g){b[g]=b[g]|| d[g]});b.google_loader_used="aa";b.google_reactive_tag_first=1===(U(L).first_tag_on_page||0);Wi(164,function(){Co(d,b,a,c)})}} function Xo(a,b,c){var d=b.google_reactive_ads_config,e="string"===typeof a.className&&RegExp("(\\W|^)adsbygoogle-noablate(\\W|$)").test(a.className),f=ul(c);if(f&&f.Fa&&"on"!=b.google_adtest&&!e){e=ni(a,c);var g=Vf(c).clientHeight;if(!f.qa||f.qa&&((0==g?null:e/g)||0)>=f.qa)return a.className+=" adsbygoogle-ablated-ad-slot",c=c.google_sv_map=c.google_sv_map||{},d=za(a),b.google_element_uid=d,c[b.google_element_uid]=b,a.setAttribute("google_element_uid",d),"slot"==f.tb&&(null!==Zc(a.getAttribute("width"))&& a.setAttribute("width",0),null!==Zc(a.getAttribute("height"))&&a.setAttribute("height",0),a.style.width="0px",a.style.height="0px"),!0}if((f=Nc(a,c))&&"none"==f.display&&!("on"==b.google_adtest||0<b.google_reactive_ad_format||d))return c.document.createComment&&a.appendChild(c.document.createComment("No ad requested because of display:none on the adsbygoogle tag")),!0;a=null==b.google_pgb_reactive||3===b.google_pgb_reactive;return 1!==b.google_reactive_ad_format&&8!==b.google_reactive_ad_format|| !a?!1:(w.console&&w.console.warn("Adsbygoogle tag with data-reactive-ad-format="+b.google_reactive_ad_format+" is deprecated. Check out page-level ads at https://www.google.com/adsense"),!0)}function Yo(a){var b=document.getElementsByTagName("INS");for(var c=0,d=b[c];c<b.length;d=b[++c]){var e=d;if(Uo(e)&&"reserved"!=e.dataset.adsbygoogleStatus&&(!a||d.id==a))return d}return null} function Zo(a,b,c){if(a&&a.shift)for(var d=20;0<a.length&&0<d;){try{$o(a.shift(),b,c)}catch(e){setTimeout(function(){throw e;})}--d}}function ap(){var a=Mc("INS");a.className="adsbygoogle";a.className+=" adsbygoogle-noablate";bd(a);return a} function bp(a,b){var c={};Sc(Rf,function(f,g){!1===a.enable_page_level_ads?c[g]=!1:a.hasOwnProperty(g)&&(c[g]=a[g])});ya(a.enable_page_level_ads)&&(c.page_level_pubvars=a.enable_page_level_ads);var d=ap();hd.body.appendChild(d);var e={};e=(e.google_reactive_ads_config=c,e.google_ad_client=a.google_ad_client,e);e.google_pause_ad_requests=!!U(L).pause_ad_requests;Vo(d,e,b)} function cp(a,b){function c(){return bp(a,b)}Uf(w).wasPlaTagProcessed=!0;var d=w.document;if(d.body||"complete"==d.readyState||"interactive"==d.readyState)c();else{var e=wc(Xi(191,c));xc(d,"DOMContentLoaded",e);(new w.MutationObserver(function(f,g){d.body&&(e(),g.disconnect())})).observe(d,{childList:!0,subtree:!0})}} function $o(a,b,c){var d={};Wi(165,function(){dp(a,d,b,c)},function(e){e.client=e.client||d.google_ad_client||a.google_ad_client;e.slotname=e.slotname||d.google_ad_slot;e.tag_origin=e.tag_origin||d.google_tag_origin})}function ep(a){delete a.google_checked_head;Sc(a,function(b,c){Ei[c]||(delete a[c],w.console.warn("AdSense head tag doesn't support "+c.replace("google","data").replace(/_/g,"-")+" attribute."))})} function fp(a,b){var c=L.document.querySelector('script[src*="/pagead/js/adsbygoogle.js?client="]:not([data-checked-head])')||L.document.querySelector('script[src*="/pagead/js/adsbygoogle.js"][data-ad-client]:not([data-checked-head])');if(c){c.setAttribute("data-checked-head","true");var d=U(window);if(d.head_tag_slot_vars)gp(c);else{var e={};Io(c,e);ep(e);var f=$b(e);d.head_tag_slot_vars=f;c={google_ad_client:e.google_ad_client,enable_page_level_ads:e};L.adsbygoogle||(L.adsbygoogle=[]);d=L.adsbygoogle; d.loaded?d.push(c):d.splice(0,0,c);var g;e.google_adbreak_test||(null==(g=Ib(b,Fk,13,Uk))?0:D(g,3))&&P(jf)?hp(f,a):No(function(){hp(f,a)})}}}function gp(a){var b=U(window).head_tag_slot_vars,c=a.getAttribute("src")||"";if((a=Gc(c)||a.getAttribute("data-ad-client")||"")&&a!==b.google_ad_client)throw new T("Warning: Do not add multiple property codes with AdSense tag to avoid seeing unexpected behavior. These codes were found on the page "+a+", "+b.google_ad_client);} function ip(a){if("object"===typeof a&&null!=a){if("string"===typeof a.type)return 2;if("string"===typeof a.sound||"string"===typeof a.preloadAdBreaks)return 3}return 0} function dp(a,b,c,d){if(null==a)throw new T("push() called with no parameters.");14===Cb(d,Uk)&&jp(a,wb(Tk(d),1),I(d,2));var e=ip(a);if(0!==e)P(af)&&(d=xl(),d.first_slotcar_request_processing_time||(d.first_slotcar_request_processing_time=Date.now(),d.adsbygoogle_execution_start_time=Ia)),null==Qo?(kp(a),Ro.push(a)):3===e?Wi(787,function(){Qo.handleAdConfig(a)}):Zi(730,Qo.handleAdBreak(a));else{Ia=(new Date).getTime();wo(c,d,lp(a));mp();a:{if(void 0!=a.enable_page_level_ads){if("string"===typeof a.google_ad_client){e= !0;break a}throw new T("'google_ad_client' is missing from the tag config.");}e=!1}if(e)np(a,d);else if((e=a.params)&&Sc(e,function(g,h){b[h]=g}),"js"===b.google_ad_output)console.warn("Ads with google_ad_output='js' have been deprecated and no longer work. Contact your AdSense account manager or switch to standard AdSense ads.");else{e=op(a.element);Io(e,b);c=U(w).head_tag_slot_vars||{};Sc(c,function(g,h){b.hasOwnProperty(h)||(b[h]=g)});if(e.hasAttribute("data-require-head")&&!U(w).head_tag_slot_vars)throw new T("AdSense head tag is missing. AdSense body tags don't work without the head tag. You can copy the head tag from your account on https://adsense.com."); if(!b.google_ad_client)throw new T("Ad client is missing from the slot.");b.google_apsail=dl(b.google_ad_client);var f=(c=0===(U(L).first_tag_on_page||0)&&Fl(b))&&Gl(c);c&&!f&&(np(c,d),U(L).skip_next_reactive_tag=!0);0===(U(L).first_tag_on_page||0)&&(U(L).first_tag_on_page=2);b.google_pause_ad_requests=!!U(L).pause_ad_requests;Vo(e,b,d);c&&f&&pp(c)}}}var qp=!1;function jp(a,b,c){P(Ye)&&!qp&&(qp=!0,a=lp(a)||qn(L),Yi("predictive_abg",{a_c:a,p_c:b,b_v:c},.01))} function lp(a){return a.google_ad_client?a.google_ad_client:(a=a.params)&&a.google_ad_client?a.google_ad_client:""}function mp(){if(P(Re)){var a=ul(L);if(!(a=a&&a.Fa)){try{var b=L.localStorage}catch(c){b=null}b=b?zj(b):null;a=!(b&&Ck(b)&&b)}a||vl(L,1)}}function pp(a){gd(function(){Uf(w).wasPlaTagProcessed||w.adsbygoogle&&w.adsbygoogle.push(a)})} function np(a,b){if(U(L).skip_next_reactive_tag)U(L).skip_next_reactive_tag=!1;else{0===(U(L).first_tag_on_page||0)&&(U(L).first_tag_on_page=1);if(a.tag_partner){var c=a.tag_partner,d=U(w);d.tag_partners=d.tag_partners||[];d.tag_partners.push(c)}U(L).ama_ran_on_page||Il(new Hl(a,b));cp(a,b)}} function op(a){if(a){if(!Uo(a)&&(a.id?a=Yo(a.id):a=null,!a))throw new T("'element' has already been filled.");if(!("innerHTML"in a))throw new T("'element' is not a good DOM element.");}else if(a=Yo(),!a)throw new T("All ins elements in the DOM with class=adsbygoogle already have ads in them.");return a} function rp(){var a=new Oj(L),b=new Oo(L),c=new Po(L),d=L.__cmp?1:0;a=Pj(a)?1:0;var e,f;(f="function"===typeof(null==(e=b.i)?void 0:e.__uspapi))||(b.h?b=b.h:(b.h=$c(b.i,"__uspapiLocator"),b=b.h),f=null!=b);c.l||(c.i||(c.i=c.h.googlefc?c.h:$c(c.h,"googlefcPresent")),c.l=!0);Yi("cmpMet",{tcfv1:d,tcfv2:a,usp:f?1:0,fc:c.i?1:0,ptt:9},Q(ye))}function sp(a){a={value:D(a,16)};var b=.01;Q(Te)&&(a.eid=Q(Te),b=1);a.frequency=b;Yi("new_abg_tag",a,b)}function tp(a){Dj().S[Fj(26)]=!!Number(a)} function up(a){Number(a)?U(L).pause_ad_requests=!0:(U(L).pause_ad_requests=!1,a=function(){if(!U(L).pause_ad_requests){var b=void 0===b?{}:b;if("function"===typeof window.CustomEvent)var c=new CustomEvent("adsbygoogle-pub-unpause-ad-requests-event",b);else c=document.createEvent("CustomEvent"),c.initCustomEvent("adsbygoogle-pub-unpause-ad-requests-event",!!b.bubbles,!!b.cancelable,b.detail);L.dispatchEvent(c)}},w.setTimeout(a,0),w.setTimeout(a,1E3))} function vp(a){Yi("adsenseGfpKnob",{value:a,ptt:9},.1);switch(a){case 0:case 2:a=!0;break;case 1:a=!1;break;default:throw Error("Illegal value of cookieOptions: "+a);}L._gfp_a_=a}function wp(a){a&&a.call&&"function"===typeof a&&window.setTimeout(a,0)} function hp(a,b){b=Dl(ec(hc(qc(b.sb).toString()),zf()?{bust:zf()}:{})).then(function(c){null==Qo&&(c.init(a),Qo=c,xp())});Zi(723,b);r(b,"finally").call(b,function(){Ro.length=0;Yi("slotcar",{event:"api_ld",time:Date.now()-Ia,time_pr:Date.now()-To})})} function xp(){for(var a=u(r(So,"keys").call(So)),b=a.next();!b.done;b=a.next()){b=b.value;var c=So.get(b);-1!==c&&(w.clearTimeout(c),So.delete(b))}a={};for(b=0;b<Ro.length;a={fa:a.fa,ba:a.ba},b++)So.has(b)||(a.ba=Ro[b],a.fa=ip(a.ba),Wi(723,function(d){return function(){3===d.fa?Qo.handleAdConfig(d.ba):2===d.fa&&Zi(730,Qo.handleAdBreakBeforeReady(d.ba))}}(a)))} function kp(a){var b=Ro.length;if(2===ip(a)&&"preroll"===a.type&&null!=a.adBreakDone){-1===To&&(To=Date.now());var c=w.setTimeout(function(){try{(0,a.adBreakDone)({breakType:"preroll",breakName:a.name,breakFormat:"preroll",breakStatus:"timeout"}),So.set(b,-1),Yi("slotcar",{event:"pr_to",source:"adsbygoogle"})}catch(d){console.error("[Ad Placement API] adBreakDone callback threw an error:",d instanceof Error?d:Error(String(d)))}},1E3*Q(kf));So.set(b,c)}} function yp(){if(P(Ne)&&!P(Me)){var a=L.document,b=a.createElement("LINK"),c=nd(Ml);if(c instanceof cc||c instanceof mc)b.href=sc(c);else{if(-1===tc.indexOf("stylesheet"))throw Error('TrustedResourceUrl href attribute required with rel="stylesheet"');b.href=rc(c)}b.rel="stylesheet";a.head.appendChild(b)}};(function(a,b,c,d){d=void 0===d?function(){}:d;Ui.Ua($i);Wi(166,function(){var e=sn(b);jn(I(e,2));Xk(D(e,6));d();kd(16,[1,e.toJSON()]);var f=md(ld(L))||L,g=c(kn({eb:a,nb:I(e,2)}),e);P(cf)&&al(f,e);om(f,e,null===L.document.currentScript?1:Ol(g.ub));Mo();if((!Na()||0<=Ka(Qa(),11))&&(null==(L.Prototype||{}).Version||!P(We))){Vi(P(qf));Go(e);ok();try{Mn()}catch(q){}Fo();fp(g,e);f=window;var h=f.adsbygoogle;if(!h||!h.loaded){if(P(Se)&&!D(e,16))try{if(L.document.querySelector('script[src*="/pagead/js/adsbygoogle.js?client="]'))return}catch(q){}yp(); sp(e);Q(ye)&&rp();var k={push:function(q){$o(q,g,e)},loaded:!0};try{Object.defineProperty(k,"requestNonPersonalizedAds",{set:tp}),Object.defineProperty(k,"pauseAdRequests",{set:up}),Object.defineProperty(k,"cookieOptions",{set:vp}),Object.defineProperty(k,"onload",{set:wp})}catch(q){}if(h)for(var l=u(["requestNonPersonalizedAds","pauseAdRequests","cookieOptions"]),m=l.next();!m.done;m=l.next())m=m.value,void 0!==h[m]&&(k[m]=h[m]);"_gfp_a_"in window||(window._gfp_a_=!0);Zo(h,g,e);f.adsbygoogle=k;h&& (k.onload=h.onload);(f=Bo(g))&&document.documentElement.appendChild(f)}}})})("m202204040101",rn,function(a,b){var c=2012<C(b,1,0)?"_fy"+C(b,1,0):"",d=I(b,3),e=I(b,2);b=nd(dn,a,c);d=nd(en,e,d);return{sb:b,qb:nd(fn,a,c),ob:nd(gn,a,c),pb:nd(hn,a,c),vb:d,ub:/^(?:https?:)?\/\/(?:pagead2\.googlesyndication\.com|securepubads\.g\.doubleclick\.net)\/pagead\/(?:js\/)?(?:show_ads|adsbygoogle)\.js(?:[?#].*)?$/}}); }).call(this,"[2019,\"r20220406\",\"r20190131\",null,null,null,null,\".google.co.uz\",null,null,null,[[[1082,null,null,[1]],[null,62,null,[null,0.001]],[383,null,null,[1]],[null,1130,null,[null,100]],[null,1126,null,[null,5000]],[1132,null,null,[1]],[1131,null,null,[1]],[null,1142,null,[null,2]],[null,1165,null,[null,1000]],[null,1114,null,[null,1]],[null,1116,null,[null,300]],[null,1117,null,[null,100]],[null,1115,null,[null,1]],[null,1159,null,[null,500]],[1145,null,null,[1]],[1021,null,null,[1]],[null,66,null,[null,-1]],[null,65,null,[null,-1]],[1087,null,null,[1]],[1053,null,null,[1]],[1100,null,null,[1]],[1102,null,null,[1]],[1149,null,null,[1]],[null,1072,null,[null,0.75]],[1101,null,null,[1]],[1036,null,null,[1]],[null,1085,null,[null,5]],[null,63,null,[null,30]],[null,1080,null,[null,5]],[1054,null,null,[1]],[null,1027,null,[null,10]],[null,57,null,[null,120]],[null,1079,null,[null,5]],[null,1050,null,[null,30]],[null,58,null,[null,120]],[381914117,null,null,[1]],[null,null,null,[null,null,null,[\"A8FHS1NmdCwGqD9DwOicnHHY+y27kdWfxKa0YHSGDfv0CSpDKRHTQdQmZVPDUdaFWUsxdgVxlwAd6o+dhJykPA0AAACWeyJvcmlnaW4iOiJodHRwczovL2RvdWJsZWNsaWNrLm5ldDo0NDMiLCJmZWF0dXJlIjoiQ29udmVyc2lvbk1lYXN1cmVtZW50IiwiZXhwaXJ5IjoxNjQzMTU1MTk5LCJpc1N1YmRvbWFpbiI6dHJ1ZSwiaXNUaGlyZFBhcnR5Ijp0cnVlLCJ1c2FnZSI6InN1YnNldCJ9\",\"A8zdXi6dr1hwXEUjQrYiyYQGlU3557y5QWDnN0Lwgj9ePt66XMEvNkVWOEOWPd7TP9sBQ25X0Q15Lr1Nn4oGFQkAAACceyJvcmlnaW4iOiJodHRwczovL2dvb2dsZXN5bmRpY2F0aW9uLmNvbTo0NDMiLCJmZWF0dXJlIjoiQ29udmVyc2lvbk1lYXN1cmVtZW50IiwiZXhwaXJ5IjoxNjQzMTU1MTk5LCJpc1N1YmRvbWFpbiI6dHJ1ZSwiaXNUaGlyZFBhcnR5Ijp0cnVlLCJ1c2FnZSI6InN1YnNldCJ9\",\"A4\/Htern2udN9w3yJK9QgWQxQFruxOXsXL7cW60DyCl0EZFGCSme\/J33Q\/WzF7bBkVvhEWDlcBiUyZaim5CpFQwAAACceyJvcmlnaW4iOiJodHRwczovL2dvb2dsZXRhZ3NlcnZpY2VzLmNvbTo0NDMiLCJmZWF0dXJlIjoiQ29udmVyc2lvbk1lYXN1cmVtZW50IiwiZXhwaXJ5IjoxNjQzMTU1MTk5LCJpc1N1YmRvbWFpbiI6dHJ1ZSwiaXNUaGlyZFBhcnR5Ijp0cnVlLCJ1c2FnZSI6InN1YnNldCJ9\"]],null,1934],[1953,null,null,[1]],[1947,null,null,[1]],[434462125,null,null,[1]],[1938,null,null,[1]],[1948,null,null,[1]],[392736476,null,null,[1]],[null,null,null,[null,null,null,[\"AxujKG9INjsZ8\/gUq8+dTruNvk7RjZQ1oFhhgQbcTJKDnZfbzSTE81wvC2Hzaf3TW4avA76LTZEMdiedF1vIbA4AAABueyJvcmlnaW4iOiJodHRwczovL2ltYXNkay5nb29nbGVhcGlzLmNvbTo0NDMiLCJmZWF0dXJlIjoiVHJ1c3RUb2tlbnMiLCJleHBpcnkiOjE2NTI3NzQ0MDAsImlzVGhpcmRQYXJ0eSI6dHJ1ZX0=\",\"Azuce85ORtSnWe1MZDTv68qpaW3iHyfL9YbLRy0cwcCZwVnePnOmkUJlG8HGikmOwhZU22dElCcfrfX2HhrBPAkAAAB7eyJvcmlnaW4iOiJodHRwczovL2RvdWJsZWNsaWNrLm5ldDo0NDMiLCJmZWF0dXJlIjoiVHJ1c3RUb2tlbnMiLCJleHBpcnkiOjE2NTI3NzQ0MDAsImlzU3ViZG9tYWluIjp0cnVlLCJpc1RoaXJkUGFydHkiOnRydWV9\",\"A16nvcdeoOAqrJcmjLRpl1I6f3McDD8EfofAYTt\/P\/H4\/AWwB99nxiPp6kA0fXoiZav908Z8etuL16laFPUdfQsAAACBeyJvcmlnaW4iOiJodHRwczovL2dvb2dsZXRhZ3NlcnZpY2VzLmNvbTo0NDMiLCJmZWF0dXJlIjoiVHJ1c3RUb2tlbnMiLCJleHBpcnkiOjE2NTI3NzQ0MDAsImlzU3ViZG9tYWluIjp0cnVlLCJpc1RoaXJkUGFydHkiOnRydWV9\",\"AxBHdr0J44vFBQtZUqX9sjiqf5yWZ\/OcHRcRMN3H9TH+t90V\/j3ENW6C8+igBZFXMJ7G3Pr8Dd13632aLng42wgAAACBeyJvcmlnaW4iOiJodHRwczovL2dvb2dsZXN5bmRpY2F0aW9uLmNvbTo0NDMiLCJmZWF0dXJlIjoiVHJ1c3RUb2tlbnMiLCJleHBpcnkiOjE2NTI3NzQ0MDAsImlzU3ViZG9tYWluIjp0cnVlLCJpc1RoaXJkUGFydHkiOnRydWV9\",\"A88BWHFjcawUfKU3lIejLoryXoyjooBXLgWmGh+hNcqMK44cugvsI5YZbNarYvi3roc1fYbHA1AVbhAtuHZflgEAAAB2eyJvcmlnaW4iOiJodHRwczovL2dvb2dsZS5jb206NDQzIiwiZmVhdHVyZSI6IlRydXN0VG9rZW5zIiwiZXhwaXJ5IjoxNjUyNzc0NDAwLCJpc1N1YmRvbWFpbiI6dHJ1ZSwiaXNUaGlyZFBhcnR5Ijp0cnVlfQ==\"]],null,1932],[null,397907552,null,[null,500]],[432938498,null,null,[1]]],[[10,[[1,[[21066108],[21066109,[[316,null,null,[1]]]]],null,null,null,34,18,1],[1,[[21066110],[21066111]],null,null,null,34,18,1],[1,[[42530528],[42530529,[[368,null,null,[1]]]],[42530530,[[369,null,null,[1]],[368,null,null,[1]]]]]],[1,[[42531496],[42531497,[[1161,null,null,[1]]]]]],[1,[[42531513],[42531514,[[316,null,null,[1]]]]]],[1,[[44719338],[44719339,[[334,null,null,[1]],[null,54,null,[null,100]],[null,66,null,[null,10]],[null,65,null,[null,1000]]]]]],[200,[[44760474],[44760475,[[1129,null,null,[1]]]]]],[10,[[44760911],[44760912,[[1160,null,null,[1]]]]]],[100,[[44761043],[44761044]]],[1,[[44752536,[[1122,null,null,[1]],[1033,null,null,[1]]]],[44753656]]],[null,[[44755592],[44755593,[[1122,null,null,[1]],[1033,null,null,[1]]]],[44755594,[[1122,null,null,[1]],[1033,null,null,[1]]]],[44755653,[[1122,null,null,[1]],[1033,null,null,[1]]]]]],[10,[[44762453],[44762454,[[1122,null,null,[1]],[1033,null,null,[1]]]]]],[20,[[182982000,[[218,null,null,[1]]],[1,[[12,null,null,null,2,null,\"\\\\.wiki(dogs|how)(-fun)?\\\\.\"]]]],[182982100,[[217,null,null,[1]]],[1,[[12,null,null,null,2,null,\"\\\\.wiki(dogs|how)(-fun)?\\\\.\"]]]]],null,null,null,36,8,1],[20,[[182982200,null,[1,[[12,null,null,null,2,null,\"\\\\.wiki(dogs|how)(-fun)?\\\\.\"]]]],[182982300,null,[1,[[12,null,null,null,2,null,\"\\\\.wiki(dogs|how)(-fun)?\\\\.\"]]]]],null,null,null,36,8,1],[10,[[182984000,null,[4,null,23,null,null,null,null,[\"1\"]]],[182984100,[[218,null,null,[1]]],[4,null,23,null,null,null,null,[\"1\"]]]],null,null,null,36,10,101],[10,[[182984200,null,[4,null,23,null,null,null,null,[\"1\"]]],[182984300,null,[4,null,23,null,null,null,null,[\"1\"]]]],null,null,null,36,10,101],[10,[[21066428],[21066429]]],[10,[[21066430],[21066431],[21066432],[21066433]],null,null,null,44,22],[10,[[21066434],[21066435]],null,null,null,44,null,500],[10,[[31065342],[31065343,[[1147,null,null,[1]]]]]],[50,[[31065544],[31065545,[[1154,null,null,[1]]]]]],[50,[[31065741],[31065742,[[1134,null,null,[1]]]]]],[1,[[31065944,[[null,1103,null,[null,31065944]],[1121,null,null,[1]],[null,1119,null,[null,300]]]],[31065945,[[null,1103,null,[null,31065945]],[1121,null,null,[1]],[1143,null,null,[1]],[null,1119,null,[null,300]]]],[31065946,[[null,1103,null,[null,31065946]],[1121,null,null,[1]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[31065950,[[null,1103,null,[null,31065950]],[null,1114,null,[null,0.9]],[null,1112,null,[null,5]],[null,1113,null,[null,5]],[null,1104,null,[null,100]],[null,1115,null,[null,0.8]],[1121,null,null,[1]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[31065951,[[null,1103,null,[null,31065951]],[null,1114,null,[null,0.9]],[null,1110,null,[null,1]],[null,1104,null,[null,100]],[null,1115,null,[null,0.8]],[1121,null,null,[1]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[31065952,[[null,1103,null,[null,31065952]],[null,1114,null,[null,0.9]],[null,1110,null,[null,5]],[null,1104,null,[null,100]],[null,1115,null,[null,0.8]],[1121,null,null,[1]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[31065953,[[null,1103,null,[null,31065953]],[null,1114,null,[null,0.9]],[null,1110,null,[null,5]],[null,1111,null,[null,5]],[null,1112,null,[null,5]],[null,1113,null,[null,5]],[null,1104,null,[null,100]],[null,1115,null,[null,0.8]],[1121,null,null,[1]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[44762492,[[null,1103,null,[null,44762492]],[null,1114,null,[null,0.9]],[null,1104,null,[null,100]],[null,1106,null,[null,10]],[null,1107,null,[null,10]],[null,1105,null,[null,10]],[null,1115,null,[null,-1]],[1121,null,null,[1]],[null,1119,null,[null,300]],[1120,null,null,[1]]]]],[6,null,null,3,null,2],49],[1,[[31066496,[[null,1103,null,[null,31066496]],[1121,null,null,[1]],[null,1119,null,[null,300]]]],[31066497,[[null,1158,null,[null,45]],[null,1157,null,[null,400]],[null,1103,null,[null,31066497]],[null,1114,null,[null,-1]],[null,1104,null,[null,100]],[null,1106,null,[null,10]],[null,1107,null,[null,10]],[null,1105,null,[null,10]],[null,1115,null,[null,-1]],[1121,null,null,[1]],[null,1119,null,[null,300]],[1162,null,null,[1]],[1155,null,null,[1]],[1120,null,null,[1]]]]],null,49],[1000,[[31067051,[[null,null,14,[null,null,\"31067051\"]]],[6,null,null,null,6,null,\"31067051\"]],[31067052,[[null,null,14,[null,null,\"31067052\"]]],[6,null,null,null,6,null,\"31067052\"]]],[4,null,55]],[1000,[[31067063,[[null,null,14,[null,null,\"31067063\"]]],[6,null,null,null,6,null,\"31067063\"]],[31067064,[[null,null,14,[null,null,\"31067064\"]]],[6,null,null,null,6,null,\"31067064\"]]],[4,null,55]],[10,[[31067067],[31067068,[[1148,null,null,[1]]]]]],[1000,[[31067083,[[null,null,14,[null,null,\"31067083\"]]],[6,null,null,null,6,null,\"31067083\"]],[31067084,[[null,null,14,[null,null,\"31067084\"]]],[6,null,null,null,6,null,\"31067084\"]]],[4,null,55]],[1,[[44736076],[44736077,[[null,1046,null,[null,0.1]]]]]],[1,[[44761631,[[null,1103,null,[null,44761631]]]],[44761632,[[null,1103,null,[null,44761632]],[1143,null,null,[1]]]],[44761633,[[null,1142,null,[null,2]],[null,1103,null,[null,44761633]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[44761634,[[null,1142,null,[null,2]],[null,1103,null,[null,44761634]],[null,1114,null,[null,0.9]],[null,1104,null,[null,100]],[null,1115,null,[null,0.8]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[44761635,[[null,1142,null,[null,2]],[null,1103,null,[null,44761635]],[null,1114,null,[null,0.9]],[null,1106,null,[null,10]],[null,1115,null,[null,0.8]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[44761636,[[null,1142,null,[null,2]],[null,1103,null,[null,44761636]],[null,1114,null,[null,0.9]],[null,1107,null,[null,10]],[null,1115,null,[null,0.8]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[44761637,[[null,1142,null,[null,2]],[null,1103,null,[null,44761637]],[null,1114,null,[null,0.9]],[null,1105,null,[null,10]],[null,1115,null,[null,0.8]],[null,1119,null,[null,300]],[1120,null,null,[1]]]],[44762110,[[null,1142,null,[null,2]],[null,1103,null,[null,44762110]],[null,1114,null,[null,0.9]],[null,1104,null,[null,100]],[null,1115,null,[null,-1]],[null,1119,null,[null,300]],[1120,null,null,[1]]]]],[6,null,null,3,null,2],49],[500,[[44761838,[[null,1142,null,[null,2]],[null,1103,null,[null,44761838]],[null,1114,null,[null,0.9]],[null,1104,null,[null,100]],[null,1115,null,[null,-1]],[null,1119,null,[null,300]],[1120,null,null,[1]]]]],[2,[[6,null,null,3,null,2],[12,null,null,null,2,null,\"smitmehta\\\\.com\/\"]]],49],[null,[[44762338],[44762339,[[380254521,null,null,[1]]]]],[1,[[4,null,63]]],null,null,56],[150,[[31061760],[31063913,[[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,200],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,500]]]]]],[31065341,[[1150,null,null,[1]],[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,200],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,500]]]]]]],[3,[[4,null,8,null,null,null,null,[\"gmaSdk.getQueryInfo\"]],[4,null,8,null,null,null,null,[\"webkit.messageHandlers.getGmaQueryInfo.postMessage\"]],[4,null,8,null,null,null,null,[\"webkit.messageHandlers.getGmaSig.postMessage\"]]]],15],[50,[[31061761,[[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,200],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,500]]]]]],[31062202,[[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,200],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,500]]]]]],[31063912],[44756455,[[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,200],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,500]]]]]]],[3,[[4,null,8,null,null,null,null,[\"gmaSdk.getQueryInfo\"]],[4,null,8,null,null,null,null,[\"webkit.messageHandlers.getGmaQueryInfo.postMessage\"]],[4,null,8,null,null,null,null,[\"webkit.messageHandlers.getGmaSig.postMessage\"]]]],15],[null,[[31063202,[[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,200],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,500]]]]]]],[3,[[4,null,8,null,null,null,null,[\"gmaSdk.getQueryInfo\"]],[4,null,8,null,null,null,null,[\"webkit.messageHandlers.getGmaQueryInfo.postMessage\"]],[4,null,8,null,null,null,null,[\"webkit.messageHandlers.getGmaSig.postMessage\"]]]],15],[null,[[44753753,[[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,200],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,500]]]]]]],[3,[[4,null,8,null,null,null,null,[\"gmaSdk.getQueryInfo\"]],[4,null,8,null,null,null,null,[\"webkit.messageHandlers.getGmaQueryInfo.postMessage\"]],[4,null,8,null,null,null,null,[\"webkit.messageHandlers.getGmaSig.postMessage\"]]]],15]]],[20,[[50,[[31062930],[31062931,[[380025941,null,null,[1]]]]],null,null,null,null,null,101,null,102]]],[13,[[10,[[44759847],[44759848,[[1947,null,null,[]]]]]],[10,[[44759849],[44759850]]],[1,[[31065824],[31065825,[[424117738,null,null,[1]]]]]],[10,[[31066184],[31066185,[[436251930,null,null,[1]]]]]],[1000,[[21067496]],[4,null,9,null,null,null,null,[\"document.hasTrustToken\"]]],[1000,[[31060475,null,[2,[[1,[[4,null,9,null,null,null,null,[\"window.PeriodicSyncManager\"]]]],[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]]]]]]],[500,[[31061692],[31061693,[[77,null,null,[1]],[78,null,null,[1]],[85,null,null,[1]],[80,null,null,[1]],[76,null,null,[1]]]]],[4,null,6,null,null,null,null,[\"31061691\"]]],[1,[[31062890],[31062891,[[397841828,null,null,[1]]]]]],[1,[[31062946]],[4,null,27,null,null,null,null,[\"document.prerendering\"]]],[1,[[31062947]],[1,[[4,null,27,null,null,null,null,[\"document.prerendering\"]]]]],[50,[[31064018],[31064019,[[1961,null,null,[1]]]]]],[1,[[31065981,null,[2,[[6,null,null,3,null,0],[12,null,null,null,4,null,\"Chrome\/(9[23456789]|\\\\d{3,})\",[\"navigator.userAgent\"]],[4,null,27,null,null,null,null,[\"crossOriginIsolated\"]]]]]]]]],[11,[[10,[[44760494],[44760495,[[1957,null,null,[1]]]]],null,48],[1,[[44760496],[44760497,[[1957,null,null,[1]]]],[44760498,[[1957,null,null,[1]]]]],null,48],[2,[[44761535],[44761536,[[1957,null,null,[1]],[1963,null,null,[1]]]],[44761537,[[1957,null,null,[1]],[1964,null,null,[1]]]],[44761538,[[1957,null,null,[1]],[1965,null,null,[1]]]],[44761539,[[1957,null,null,[1]]]]],null,48]]],[17,[[10,[[31060047]],null,null,null,44,null,900],[10,[[31060048],[31060049]],null,null,null,null,null,null,null,101],[10,[[31060566]]]]],[12,[[50,[[31061828],[31061829,[[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,200],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,500]]]],[360245597,null,null,[1]],[null,494,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]]]],[31065659,[[1150,null,null,[1]],[null,1126,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]],[null,1032,null,[null,10000]],[427841102,null,null,[1]],[360245597,null,null,[1]],[null,494,null,[null,5000],[[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[null,5500]]]]]],[31065787]],null,15],[20,[[21065724],[21065725,[[203,null,null,[1]]]]],[4,null,9,null,null,null,null,[\"LayoutShift\"]]],[50,[[31060006,null,[2,[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[12,null,null,null,4,null,\"Chrome\/(89|9\\\\d|\\\\d{3,})\",[\"navigator.userAgent\"]],[4,null,9,null,null,null,null,[\"window.PeriodicSyncManager\"]]]]],[31060007,[[1928,null,null,[1]]],[2,[[12,null,null,null,4,null,\"Android\",[\"navigator.userAgent\"]],[12,null,null,null,4,null,\"Chrome\/(89|9\\\\d|\\\\d{3,})\",[\"navigator.userAgent\"]],[4,null,9,null,null,null,null,[\"window.PeriodicSyncManager\"]]]]]],null,21],[10,[[31060032],[31060033,[[1928,null,null,[1]]]]],null,21],[10,[[31061690],[31061691,[[83,null,null,[1]],[84,null,null,[1]]]]]],[1,[[31065721],[31065722,[[432946749,null,null,[1]]]]]]]]],null,null,[0.001,\"1000\",1,\"1000\"]],[null,[]],null,null,1,\"github.com\",309779023,[44759876,44759927,44759842]]");
sinanw / Ml Classification Malicious Network TrafficThis project aims to analyze and classify a real network traffic dataset to detect malicious/benign traffic records. It compares and tunes the performance of several Machine Learning algorithms to maintain the highest accuracy and lowest False Positive/Negative rates.
m-clark / ConfusionMatrixReport various statistics stemming from a confusion matrix in a tidy fashion. 🎯
shreyash2610 / Convolutional Long Short Term Memory Based IOT Node For Violence DetectionAbstract— Violence detection has been investigated extensively in the literature. Recently, IOT based violence video surveillance is an intelligent component integrated in security system of smart buildings. Violence video detector is a specific kind of detection models that should be highly accurate to increase the model’s sensitivity and reduce the false alarm rate. This paper proposes a novel architecture of ConvLSTM model that can run on low-cost Internet of Things (IOT) device such as raspberry pi board. The paper utilized convolutional neural networks (CNNs) to learn spatial features from video’s frames that were applied to Long Short- Term Memory (LSTM) for video classification into violence/non-violence classes. A complex dataset including two public datasets: RWF-2000 and RLVS-2000 was used for model training and evaluation. The challenging video content includes crowds and chaos, small object at far distance, low resolution, and transient action. Additionally, the videos were captured in various environments such as street, prison, and schools with several human actions such as playing football, basketball, tennis, swimming and eating. The experimental results show high performance of the proposed violence detection model in terms of average metrics having an accuracy of 73.35 %, recall of 76.90 %, precision of 72.53 %, F1 score of 74.01 %, false negative rate of 23.10 %, false positive rate of 30.20 %, and AUC of 82.0 %.
lilygrace454 / Mcafee.com ActivateThe Best Antivirus and Total Protection for Mac! What's the Best Malware Protection? Malware, Spyware, and Adware Protection Antivirus programming is basic for each PC. Without it, you hazard losing your own data, your records, and even the money from your financial balance. We've tried in excess of 40 utilities to enable you to pick the best antivirus security for your PCs. Malware, Spyware, and Adware Protection Summer is practically here, and we're all anticipating a stunning get-away, be it at the shoreline, in the mountains, or even on a journey. Malware coders get-aways, as well. Theirs is an occupation, similar to some other, from numerous points of view. In any case, that doesn't mean you'll be sheltered from infections, ransomware, bots, and other malware this late spring. The malware office director timetables get-aways, much the same as any office supervisor, to ensure someone's at work, making new assaults on your gadgets and your information. Before you head out, check your antivirus membership to ensure it won't take its very own get-away soon. In case you're not ensured at this point, put introducing an antivirus on your agenda. We've tried and McAfee com activate antivirus instruments so you can pick one and unwind with no stresses. We call it antivirus, yet in truth it's far-fetched you'll get hit with a real PC infection. Malware nowadays is tied in with profiting, and there's no simple method to take advantage of spreading an infection. Ransomware and information taking Trojans are considerably more typical, as are bots that let the bot-herder lease your PC for loathsome purposes. Current antivirus utilities handle Trojans, rootkits, spyware, adware, ransomware, and that's only the tip of the iceberg. PCMag has explored in excess of 40 distinctive business antivirus utilities, and that is not notwithstanding checking the many free antivirus devices. Out of that broad field we've named fourEditors' Choice items. mcafee activate product key antivirus utilities demonstrated powerful enough to procure an astounding four-star rating nearby their progressively conventional partners. VoodooSoft VoodooShield puts together its security with respect to stifling every obscure program while the PC is in a powerless state, for example, when it's associated with the web, and furthermore acts to identify known malware. The Kure resets the PC to a known safe state on each reboot, consequently wiping out any malware. On the off chance that you have malware, one of the ten items in the graph above should deal with the issue. You may see that one item in the outline earned simply 3.5 stars. The diagram had space for one more, and of the seven 3.5-star items, the labs just focus on F-Secure and G Data. F-Secure has the additional fillip of costing the equivalent for three licenses as most items charge for only one, so it advanced into the diagram. The blurbs at the base of this article incorporate each business antivirus that earned 3.5 stars or better. www.mcafee.com/activate offer insurance past the antivirus incorporated with Windows 10; the best free antivirus utilities additionally offer more. In any case, Microsoft Windows Defender Security Center is looking somewhat better recently, with some awesome scores from free testing labs. In our grasp on tests, it demonstrated a checked improvement since our past audit, enough to at long last bring it up to three stars. Tune in to the Labs We take the outcomes announced by free antivirus testing labs in all respects truly. The basic reality that a specific merchant's item appears in the outcomes is a demonstration of positive support, of sorts. It implies the lab considered the item huge, and the merchant felt the expense of testing was advantageous. Obviously, getting great scores in the tests is additionally significant. We pursue four labs that normally discharge nitty gritty reports: SE Labs, AV-Test Institute, MRG-Effitas, and AV-Comparatives. We likewise note whether sellers have contracted with ICSA Labs and West Coast labs for affirmation. We Test Malware, Spyware, and Adware Defenses We additionally subject each item to our very own hands-on trial of malware assurance, to a limited extent to get an inclination for how the item functions. Contingent upon how altogether the item averts malware establishment, it can acquire up to 10 for malware assurance. Our malware assurance test fundamentally utilizes a similar arrangement of tests for quite a long time. To check an item's treatment of fresh out of the box new malware, we test every item utilizing 100 amazingly new malware-facilitating URLs provided by MRG-Effitas, taking note of what level of them it blocked. Items get equivalent kudos for anticipating all entrance to the vindictive URL and for clearing out the malware during download. A few items win completely outstanding evaluations from the autonomous labs, yet don't toll too in our grasp on tests. In such cases, we concede to the labs, as they carry altogether more prominent assets to their testing. Need to know more? You can dive in for a point by point portrayal of how we test security programming. Multilayered Malware Protection Antivirus items separate themselves by going past the fundamentals of on-request examining and ongoing malware insurance. Some rate URLs that you visit or that appear in indexed lists, utilizing a red-yellow-green shading coding framework. Some effectively square procedures on your framework from associating with known malware-facilitating URLs or with false (phishing) pages. Programming has imperfections, and here and there those blemishes influence your mcafee security. Judicious clients keep Windows and all projects fixed, fixing those imperfections at the earliest opportunity. The defenselessness output offered by some antivirus items can check that every vital patches are available, and even apply any that are absent. Spyware comes in numerous structures, from concealed projects that log your each keystroke to Trojans that take on the appearance of legitimate projects while mining your own information. Any antivirus should deal with spyware, alongside every other sort of malware, yet some incorporate specific segments gave to spyware insurance. You expect an antivirus to recognize and dispose of terrible projects, and to disregard great projects. Shouldn't something be said about questions, programs it can't distinguish as positive or negative? Conduct based discovery can, in principle, secure you against malware that is so new scientists have never experienced it. In any case, this isn't generally an unmixed gift. It's normal for social discovery frameworks to hail numerous harmless practices performed by genuine projects. Whitelisting is another way to deal with the issue of obscure projects. A whitelist-based security framework just permits realized great projects to run. Questions are prohibited. This mode sometimes falls short for all circumstances, however it very well may be helpful. Sandboxing gives obscure projects a chance to run, however it segregates them from full access to your framework, so they can't do lasting damage. These different added layers serve to upgrade your assurance against malware. What's the Best Malware Protection? mcafee.com/activate antivirus would it be advisable for you to pick? You have an abundance of choices. Kaspersky Anti-Virus and Bitdefender Antivirus Plus routinely take impeccable or close ideal scores from the autonomous antivirus testing labs. A solitary membership for AntiVirus Plus gives you a chance to introduce security on the majority of your Windows, Android, Mac OS, and iOS gadgets. What's more, its unordinary conduct based recognition innovation implies Webroot SecureAnywhere Antivirus is the most diminutive antivirus around. We've named these four Editors' Choice for business antivirus, however they're by all account not the only items worth thought. Peruse the surveys of our top of the line items, and afterward settle on your own choice. Note that we have assessed a lot more antivirus utilities than we could incorporate into the diagram of top items. On the off chance that your preferred programming isn't recorded there, odds are we reviewed it. The blurbs beneath incorporate each item that oversaw 3.5 stars or better. Every one of the utilities recorded in this component are Windows antivirus applications. In case you're a macOS client, don't lose hope, be that as it may; PCMag has a different gathering devoted exclusively to the best Mac antivirus programming.
flybunctious / The Game Of HogIntroduction In this project, you will develop a simulator and multiple strategies for the dice game Hog. You will need to use control statements and higher-order functions together, as described in Sections 1.2 through 1.6 of Composing Programs. In Hog, two players alternate turns trying to be the first to end a turn with at least 100 total points. On each turn, the current player chooses some number of dice to roll, up to 10. That player's score for the turn is the sum of the dice outcomes. To spice up the game, we will play with some special rules: Pig Out. If any of the dice outcomes is a 1, the current player's score for the turn is 1. Example 1: The current player rolls 7 dice, 5 of which are 1's. They score 1 point for the turn. Example 2: The current player rolls 4 dice, all of which are 3's. Since Pig Out did not occur, they score 12 points for the turn. Free Bacon. A player who chooses to roll zero dice scores one more than the largest digit in the opponent's total score. Example 1: If the opponent has 42 points, the current player gains 1 + max(4, 2) = 5 points by rolling zero dice. Example 2: If the opponent has 48 points, the current player gains 1 + max(4, 8) = 9 points by rolling zero dice. Example 3: If the opponent has 7 points, the current player gains 1 + max(0, 7) = 8 points by rolling zero dice. Swine Swap. After points for the turn are added to the current player's score, if both scores are larger than 1 and either one of the scores is a positive integer multiple of the other, then the two scores are swapped. Example 1: The current player has a total score of 37 and the opponent has 92. The current player rolls two dice that total 9. The opponent's score (92) is exactly twice the player's new total score (46). These scores are swapped! The current player now has 92 points and the opponent has 46. The turn ends. Example 2: The current player has 91 and the opponent has 37. The current player rolls five dice that total 20. The current player has 111, which is 3 times 37, so the scores are swapped. The opponent ends the turn with 111 and wins the game. Download starter files To get started, download all of the project code as a zip archive. You only have to make changes to hog.py. hog.py: A starter implementation of Hog dice.py: Functions for rolling dice hog_gui.py: A graphical user interface for Hog ucb.py: Utility functions for CS 61A ok: CS 61A autograder tests: A directory of tests used by ok images: A directory of images used by hog_gui.py Logistics This is a 2-week project. This is a solo project, so you will complete this project without a partner. You should not share your code with any other students, or copy from anyone else's solutions. Remember that you can earn an additional bonus point by submitting the project at least 24 hours before the deadline. The project is worth 20 points. 18 points are assigned for correctness, and 2 points for the overall composition of your program. You will turn in the following files: hog.py You do not need to modify or turn in any other files to complete the project. To submit the project, run the following command: python3 ok --submit You will be able to view your submissions on the Ok dashboard. For the functions that we ask you to complete, there may be some initial code that we provide. If you would rather not use that code, feel free to delete it and start from scratch. You may also add new function definitions as you see fit. However, please do not modify any other functions. Doing so may result in your code failing our autograder tests. Also, please do not change any function signatures (names, argument order, or number of arguments). Testing Throughout this project, you should be testing the correctness of your code. It is good practice to test often, so that it is easy to isolate any problems. However, you should not be testing too often, to allow yourself time to think through problems. We have provided an autograder called ok to help you with testing your code and tracking your progress. The first time you run the autograder, you will be asked to log in with your Ok account using your web browser. Please do so. Each time you run ok, it will back up your work and progress on our servers. The primary purpose of ok is to test your implementations, but there are two things you should be aware of. First, some of the test cases are locked. To unlock tests, run the following command from your terminal: python3 ok -u This command will start an interactive prompt that looks like: ===================================================================== Assignment: The Game of Hog Ok, version ... ===================================================================== ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Unlocking tests At each "? ", type what you would expect the output to be. Type exit() to quit --------------------------------------------------------------------- Question 0 > Suite 1 > Case 1 (cases remaining: 1) >>> Code here ? At the ?, you can type what you expect the output to be. If you are correct, then this test case will be available the next time you run the autograder. The idea is to understand conceptually what your program should do first, before you start writing any code. Once you have unlocked some tests and written some code, you can check the correctness of your program using the tests that you have unlocked: python3 ok Most of the time, you will want to focus on a particular question. Use the -q option as directed in the problems below. We recommend that you submit after you finish each problem. Only your last submission will be graded. It is also useful for us to have more backups of your code in case you run into a submission issue. The tests folder is used to store autograder tests, so do not modify it. You may lose all your unlocking progress if you do. If you need to get a fresh copy, you can download the zip archive and copy it over, but you will need to start unlocking from scratch. If you do not want us to record a backup of your work or information about your progress, use the --local option when invoking ok. With this option, no information will be sent to our course servers. Graphical User Interface A graphical user interface (GUI, for short) is provided for you. At the moment, it doesn't work because you haven't implemented the game logic. Once you complete the play function, you will be able to play a fully interactive version of Hog! In order to render the graphics, make sure you have Tkinter, Python's main graphics library, installed on your computer. Once you've done that, you can run the GUI from your terminal: python3 hog_gui.py Once you complete the project, you can play against the final strategy that you've created! python3 hog_gui.py -f Phase 1: Simulator In the first phase, you will develop a simulator for the game of Hog. Problem 0 (0 pt) The dice.py file represents dice using non-pure zero-argument functions. These functions are non-pure because they may have different return values each time they are called. The documentation of dice.py describes the two different types of dice used in the project: Dice can be fair, meaning that they produce each possible outcome with equal probability. Example: six_sided. For testing functions that use dice, deterministic test dice always cycle through a fixed sequence of values that are passed as arguments to the make_test_dice function. Before we start writing any code, let's understand the make_test_dice function by unlocking its tests. python3 ok -q 00 -u This should display a prompt that looks like this: ===================================================================== Assignment: Project 1: Hog Ok, version v1.5.2 ===================================================================== ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Unlocking tests At each "? ", type what you would expect the output to be. Type exit() to quit --------------------------------------------------------------------- Question 0 > Suite 1 > Case 1 (cases remaining: 1) >>> test_dice = make_test_dice(4, 1, 2) >>> test_dice() ? You should type in what you expect the output to be. To do so, you need to first figure out what test_dice will do, based on the description above. You can exit the unlocker by typing exit() (without quotes). Typing Ctrl-C on Windows to exit out of the unlocker has been known to cause problems, so avoid doing so. Problem 1 (2 pt) Implement the roll_dice function in hog.py. It takes two arguments: a positive integer called num_rolls giving the number of dice to roll and a dice function. It returns the number of points scored by rolling the dice that number of times in a turn: either the sum of the outcomes or 1 (Pig Out). To obtain a single outcome of a dice roll, call dice(). You should call dice() exactly num_rolls times in the body of roll_dice. Remember to call dice() exactly num_rolls times even if Pig Out happens in the middle of rolling. In this way, we correctly simulate rolling all the dice together. Checking Your Work: Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 01 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 01 If the tests don't pass, it's time to debug. You can observe the behavior of your function using Python directly. First, start the Python interpreter and load the hog.py file. python3 -i hog.py Then, you can call your roll_dice function on any number of dice you want, such as 4. >>> roll_dice(4) In most systems, you can evaluate the same expression again by pressing the up arrow or Control-P, then pressing enter or return. You should find that evaluating this call expression gives a different answer each time, since dice rolls are random. The roll_dice function has a default argument value for dice that is a random six-sided dice function. You can also use test dice that fix the outcomes of the dice in advance. For example, rolling twice when you know that the dice will come up 3 and 4 should give a total outcome of 7. >>> fixed_dice = make_test_dice(3, 4) >>> roll_dice(2, fixed_dice) 7 If you find a problem, you need to change your hog.py file, save it, quit Python, start it again, and then start evaluating expressions. Pressing the up arrow should give you access to your previous expressions, even after restarting Python. Once you think that your roll_dice function is correct, run the ok tests again. Tests like these don't prove that your program is exactly correct, but they help you build confidence that this part of your program does what you expect, so that you can trust the abstraction it defines as you proceed. Problem 2 (1 pt) Implement the free_bacon helper function that returns the number of points scored by rolling 0 dice, based on the opponent's current score. You can assume that score is less than 100. For a score less than 10, assume that the first of the two digits is 0. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 02 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 02 You can also test free_bacon interactively by entering python3 -i hog.py in the terminal and then calling free_bacon with various inputs. Problem 3 (1 pt) Implement the take_turn function, which returns the number of points scored for a turn by the current player. Your implementation should call roll_dice when possible. You will need to implement the Free Bacon rule. You can assume that opponent_score is less than 100. Call free_bacon in your implementation of take_turn. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 03 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 03 Problem 4 (1 pt) Implement is_swap, which returns whether or not the scores should be swapped because one is an integer multiple of the other. The is_swap function takes two arguments: the player scores. It returns a boolean value to indicate whether the Swine Swap condition is met. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 04 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 04 Problem 5 (3 pt) Implement the play function, which simulates a full game of Hog. Players alternate turns, each using their respective strategy function (Player 0 uses strategy0, etc.), until one of the players reaches the goal score. When the game ends, play returns the final total scores of both players, with Player 0's score first, and Player 1's score second. Here are some hints: You should use the functions you have already written! You will need to call take_turn with all three arguments. Only call take_turn once per turn. Enforce all the special rules. You can get the number of the other player (either 0 or 1) by calling the provided function other. You can ignore the say argument to the play function for now. You will use it in Phase 2 of the project. A strategy is a function that, given a player's score and their opponent's score, returns how many dice the player wants to roll. A strategy function (such as strategy0 and strategy1) takes two arguments: scores for the current player and opposing player, which both must be non-negative integers. A strategy function returns the number of dice that the current player wants to roll in the turn. Each strategy function should be called only once per turn. Don't worry about the details of implementing strategies yet. You will develop them in Phase 3. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 05 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 05 The last test for Question 5 is a fuzz test, which checks that your play function works for a large number of different inputs. Failing this test means something is wrong, but you should look at other tests to see where the problem might be. Once you are finished, you will be able to play a graphical version of the game. We have provided a file called hog_gui.py that you can run from the terminal: python3 hog_gui.py If you don't already have Tkinter (Python's graphics library) installed, you'll need to install it first before you can run the GUI. The GUI relies on your implementation, so if you have any bugs in your code, they will be reflected in the GUI. This means you can also use the GUI as a debugging tool; however, it's better to run the tests first. Congratulations! You have finished Phase 1 of this project! Phase 2: Commentary In the second phase, you will implement commentary functions that print remarks about the game, such as, "22 points! That's the biggest gain yet for Player 1." A commentary function takes two arguments, the current score for Player 0 and the current score for Player 1. It returns another commentary function to be called on the next turn. It may also print some output as a side effect of being called. The function say_scores in hog.py is an example of a commentary function. The function announce_lead_changes is an example of a higher-order function that returns a commentary function. def say_scores(score0, score1): """A commentary function that announces the score for each player.""" print("Player 0 now has", score0, "and Player 1 now has", score1) return say_scores def announce_lead_changes(previous_leader=None): """Return a commentary function that announces lead changes. >>> f0 = announce_lead_changes() >>> f1 = f0(5, 0) Player 0 takes the lead by 5 >>> f2 = f1(5, 12) Player 1 takes the lead by 7 >>> f3 = f2(8, 12) >>> f4 = f3(8, 13) >>> f5 = f4(15, 13) Player 0 takes the lead by 2 """ def say(score0, score1): if score0 > score1: leader = 0 elif score1 > score0: leader = 1 else: leader = None if leader != None and leader != previous_leader: print('Player', leader, 'takes the lead by', abs(score0 - score1)) return announce_lead_changes(leader) return say Problem 6 (2 pt) Update your play function so that a commentary function is called at the end of each turn. say(score0, score1) should be called at the end of the first turn. Its return value (another commentary function) should be called at the end of the second turn. Each turn, a new commentary function should be called that is the return value of the previous call to a commentary function. Also implement both, a function that takes two commentary functions (f and g) and returns a new commentary function. This new commentary function returns another commentary function which calls the functions returned by calling f and g, in that order. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 06 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 06 Problem 7 (2 pt) Implement the announce_highest function, which is a higher-order function that returns a commentary function. This commentary function announces whenever a particular player gains more points in a turn than ever before. To compute the gain, it must compare the score from last turn to the score from this turn for the player of interest, which is designated by the who argument. This function must also keep track of the highest gain for the player so far. The way in which announce_highest announces is very specific, and your implementation should match the doctests provided. Notice in particular that if the gain is only 1 point, then the message includes "point" in singular form. If the gain is larger, then the message includes "points" in plural form. Use Ok to test your code: python3 ok -q 07 Hint. The announce_lead_changes function provided to you is an example of how to keep track of information using commentary functions. If you are stuck, first make sure you understand how announce_lead_changes works. When you are done, if play the game again, you will see the commentary. python3 hog_gui.py The commentary in the GUI is generated by passing the following function as the say argument to play. both(announce_highest(0), both(announce_highest(1), announce_lead_changes())) Great work! You just finished Phase 2 of the project! Phase 3: Strategies In the third phase, you will experiment with ways to improve upon the basic strategy of always rolling a fixed number of dice. First, you need to develop some tools to evaluate strategies. Problem 8 (2 pt) Implement the make_averaged function, which is a higher-order function that takes a function fn as an argument. It returns another function that takes the same number of arguments as fn (the function originally passed into make_averaged). This returned function differs from the input function in that it returns the average value of repeatedly calling fn on the same arguments. This function should call fn a total of num_samples times and return the average of the results. To implement this function, you need a new piece of Python syntax! You must write a function that accepts an arbitrary number of arguments, then calls another function using exactly those arguments. Here's how it works. Instead of listing formal parameters for a function, we write *args. To call another function using exactly those arguments, we call it again with *args. For example, >>> def printed(fn): ... def print_and_return(*args): ... result = fn(*args) ... print('Result:', result) ... return result ... return print_and_return >>> printed_pow = printed(pow) >>> printed_pow(2, 8) Result: 256 256 >>> printed_abs = printed(abs) >>> printed_abs(-10) Result: 10 10 Read the docstring for make_averaged carefully to understand how it is meant to work. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 08 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 08 Problem 9 (1 pt) Implement the max_scoring_num_rolls function, which runs an experiment to determine the number of rolls (from 1 to 10) that gives the maximum average score for a turn. Your implementation should use make_averaged and roll_dice. If two numbers of rolls are tied for the maximum average score, return the lower number. For example, if both 3 and 6 achieve a maximum average score, return 3. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 09 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 09 To run this experiment on randomized dice, call run_experiments using the -r option: python3 hog.py -r Running experiments For the remainder of this project, you can change the implementation of run_experiments as you wish. By calling average_win_rate, you can evaluate various Hog strategies. For example, change the first if False: to if True: in order to evaluate always_roll(8) against the baseline strategy of always_roll(4). You should find that it wins slightly more often than it loses, giving a win rate around 0.5. Some of the experiments may take up to a minute to run. You can always reduce the number of samples in make_averaged to speed up experiments. Problem 10 (1 pt) A strategy can take advantage of the Free Bacon rule by rolling 0 when it is most beneficial to do so. Implement bacon_strategy, which returns 0 whenever rolling 0 would give at least margin points and returns num_rolls otherwise. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 10 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 10 Once you have implemented this strategy, change run_experiments to evaluate your new strategy against the baseline. You should find that it wins more than half of the time. Problem 11 (2 pt) A strategy can also take advantage of the Swine Swap rule. The swap_strategy rolls 0 if it would cause a beneficial swap. It also returns 0 if rolling 0 would give at least margin points and would not cause a swap. Otherwise, the strategy rolls num_rolls. Before writing any code, unlock the tests to verify your understanding of the question. python3 ok -q 11 -u Once you are done unlocking, begin implementing your solution. You can check your correctness with: python3 ok -q 11 Once you have implemented this strategy, update run_experiments to evaluate your new strategy against the baseline. You should find that it gives a significant edge over always_roll(4). Optional: Problem 12 (0 pt) Implement final_strategy, which combines these ideas and any other ideas you have to achieve a high win rate against the always_roll(4) strategy. Some suggestions: swap_strategy is a good default strategy to start with. There's no point in scoring more than 100. Check whether you can win by rolling 0, 1 or 2 dice. If you are in the lead, you might take fewer risks. Try to force a beneficial swap. Choose the num_rolls and margin arguments carefully. You can check that your final strategy is valid by running Ok. python3 ok -q 12 You can also check your exact final winrate by running python3 calc.py At this point, run the entire autograder to see if there are any tests that don't pass. python3 ok Once you are satisfied, submit to Ok to complete the project. python3 ok --submit You can also play against your final strategy with the graphical user interface: python3 hog_gui.py -f The GUI will alternate which player is controlled by you. Congratulations, you have reached the end of your first CS 61A project! If you haven't already, relax and enjoy a few games of Hog with a friend.
MichaelIcaza / Localization From Image LabelsA VGG style convolutional neural network, distinguishing chromophobic renal cell carcinoma from oncocytoma. The CNN uses patch-based classification with a max-pooling fusion method to classify images. We apply dropout onto the patch-level classifications and measure the dropout rate effect on false positive and false negative rates.
prateekpr / HEART DISEASE ANALYSISLOGISTIC REGRESSION - HEART DISEASE PREDICTION Introduction World Health Organization has estimated 12 million deaths occur worldwide, every year due to Heart diseases. Half the deaths in the United States and other developed countries are due to cardio vascular diseases. The early prognosis of cardiovascular diseases can aid in making decisions on lifestyle changes in high risk patients and in turn reduce the complications. This research intends to pinpoint the most relevant/risk factors of heart disease as well as predict the overall risk using logistic regression Data Preparation Source The dataset is publically available on the Kaggle website, and it is from an ongoing cardiovascular study on residents of the town of Framingham, Massachusetts. The classification goal is to predict whether the patient has 10-year risk of future coronary heart disease (CHD).The dataset provides the patients’ information. It includes over 4,000 records and 15 attributes. Variables Each attribute is a potential risk factor. There are both demographic, behavioral and medical risk factors. Demographic: • Sex: male or female(Nominal) • Age: Age of the patient;(Continuous - Although the recorded ages have been truncated to whole numbers, the concept of age is continuous) Behavioral • Current Smoker: whether or not the patient is a current smoker (Nominal) • Cigs Per Day: the number of cigarettes that the person smoked on average in one day.(can be considered continuous as one can have any number of cigarettes, even half a cigarette.) Medical( history) • BP Meds: whether or not the patient was on blood pressure medication (Nominal) • Prevalent Stroke: whether or not the patient had previously had a stroke (Nominal) • Prevalent Hyp: whether or not the patient was hypertensive (Nominal) • Diabetes: whether or not the patient had diabetes (Nominal) Medical(current) • Tot Chol: total cholesterol level (Continuous) • Sys BP: systolic blood pressure (Continuous) • Dia BP: diastolic blood pressure (Continuous) • BMI: Body Mass Index (Continuous) • Heart Rate: heart rate (Continuous - In medical research, variables such as heart rate though in fact discrete, yet are considered continuous because of large number of possible values.) • Glucose: glucose level (Continuous) Predict variable (desired target) • 10 year risk of coronary heart disease CHD (binary: “1”, means “Yes”, “0” means “No”) Logistic Regression Logistic regression is a type of regression analysis in statistics used for prediction of outcome of a categorical dependent variable from a set of predictor or independent variables. In logistic regression the dependent variable is always binary. Logistic regression is mainly used to for prediction and also calculating the probability of success. The results above show some of the attributes with P value higher than the preferred alpha(5%) and thereby showing low statistically significant relationship with the probability of heart disease. Backward elimination approach is used here to remove those attributes with highest P-value one at a time followed by running the regression repeatedly until all attributes have P Values less than 0.05. Feature Selection: Backward elimination (P-value approach) Logistic regression equation P=eβ0+β1X1/1+eβ0+β1X1P=eβ0+β1X1/1+eβ0+β1X1 When all features plugged in: logit(p)=log(p/(1−p))=β0+β1∗Sexmale+β2∗age+β3∗cigsPerDay+β4∗totChol+β5∗sysBP+β6∗glucoselogit(p)=log(p/(1−p))=β0+β1∗Sexmale+β2∗age+β3∗cigsPerDay+β4∗totChol+β5∗sysBP+β6∗glucose Interpreting the results: Odds Ratio, Confidence Intervals and P-values • This fitted model shows that, holding all other features constant, the odds of getting diagnosed with heart disease for males (sex_male = 1)over that of females (sex_male = 0) is exp(0.5815) = 1.788687. In terms of percent change, we can say that the odds for males are 78.8% higher than the odds for females. • The coefficient for age says that, holding all others constant, we will see 7% increase in the odds of getting diagnosed with CDH for a one year increase in age since exp(0.0655) = 1.067644. • Similarly , with every extra cigarette one smokes thers is a 2% increase in the odds of CDH. • For Total cholesterol level and glucose level there is no significant change. • There is a 1.7% increase in odds for every unit increase in systolic Blood Pressure. Model Evaluation - Statistics From the above statistics it is clear that the model is highly specific than sensitive. The negative values are predicted more accurately than the positives. Predicted probabilities of 0 (No Coronary Heart Disease) and 1 ( Coronary Heart Disease: Yes) for the test data with a default classification threshold of 0.5 lower the threshold Since the model is predicting Heart disease too many type II errors is not advisable. A False Negative ( ignoring the probability of disease when there actually is one) is more dangerous than a False Positive in this case. Hence in order to increase the sensitivity, threshold can be lowered. Conclusions • All attributes selected after the elimination process show P-values lower than 5% and thereby suggesting significant role in the Heart disease prediction. • Men seem to be more susceptible to heart disease than women. Increase in age, number of cigarettes smoked per day and systolic Blood Pressure also show increasing odds of having heart disease • Total cholesterol shows no significant change in the odds of CHD. This could be due to the presence of 'good cholesterol(HDL) in the total cholesterol reading. Glucose too causes a very negligible change in odds (0.2%) • The model predicted with 0.88 accuracy. The model is more specific than sensitive. Overall model could be improved with more data
yash8005 / Facial Recognition Using Haar CascadeHuman face identification has been a testing issue in the regions of picture preparing and patter acknowledgment. Another human face location calculation by crude Haar course calculation joined with the refreshed changes are to be examined.First, pictures of individuals are handled by a crude Haar course classifier, almost without wrong human face dismissal (low rate of false negative) yet with some wrong acknowledgment (false positive). Also, to dispose of these wrongly acknowledged non-human faces, a powerless classifier in light of face skin tint histogram coordinating is connected and a dominant part of non-human countenances are evacuated. Next, another frail classifier in view of eyes identification is attached and some leftover non-human appearances are resolved and dismissed. At last, a mouth recognition task is used to the rest of the non-human appearances and the false positive rate is additionally diminished.
SUMEETRM / Covid19 AIThe 2019 novel coronavirus (COVID-19) presents several unique features. While the diagnosis is confirmed using polymerase chain reaction (PCR), infected patients with pneumonia may present on chest X-ray and computed tomography (CT) images with a pattern that is only moderately characteristic for the human eye. Bilateral multiple lobular and subsegmental areas of consolidation can be observed in COVID-19 patients. The following model aims to present a neural network aimed to detect COVID-19 cases through chest X-Rays. While a neural network to detect COVID-19 cases has been published, the following model has a much higher accuracy. Out model uses a sample of 100 COVID-19 positive cases and 100 COVID-19 negative cases, and has an accuracy of 91% . Of the true positive patients, the model had an accuracy of 100%, and of the true negative patients, the model had an accuracy of 80%. Given the fact that the data is limited, the model can be improved in the near future for rapid diagnosis of COVID-19 cases with an extremely low rate of false positives. While the model may still be in an initial phase of development, it can be put to practical use soon with enough training and accuracy.
yukubo / Sample Word2vec Skipgram# Copyright 2015 Google Inc. All Rights Reserved. # # Licensed under the Apache License, Version 2.0 (the "License"); # you may not use this file except in compliance with the License. # You may obtain a copy of the License at # # http://www.apache.org/licenses/LICENSE-2.0 # # Unless required by applicable law or agreed to in writing, software # distributed under the License is distributed on an "AS IS" BASIS, # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. # See the License for the specific language governing permissions and # limitations under the License. # ============================================================================== """Multi-threaded word2vec mini-batched skip-gram model. Trains the model described in: (Mikolov, et. al.) Efficient Estimation of Word Representations in Vector Space ICLR 2013. http://arxiv.org/abs/1301.3781 This model does traditional minibatching. The key ops used are: * placeholder for feeding in tensors for each example. * embedding_lookup for fetching rows from the embedding matrix. * sigmoid_cross_entropy_with_logits to calculate the loss. * GradientDescentOptimizer for optimizing the loss. * skipgram custom op that does input processing. """ from __future__ import absolute_import from __future__ import division from __future__ import print_function import os import sys import threading import time from six.moves import xrange # pylint: disable=redefined-builtin import numpy as np import tensorflow as tf from tensorflow.models.embedding import gen_word2vec as word2vec flags = tf.app.flags flags.DEFINE_string("save_path", None, "Directory to write the model and " "training summaries.") flags.DEFINE_string("train_data", None, "Training text file. " "E.g., unzipped file http://mattmahoney.net/dc/text8.zip.") flags.DEFINE_string( "eval_data", None, "File consisting of analogies of four tokens." "embedding 2 - embedding 1 + embedding 3 should be close " "to embedding 4." "E.g. https://word2vec.googlecode.com/svn/trunk/questions-words.txt.") flags.DEFINE_integer("embedding_size", 200, "The embedding dimension size.") flags.DEFINE_integer( "epochs_to_train", 15, "Number of epochs to train. Each epoch processes the training data once " "completely.") flags.DEFINE_float("learning_rate", 0.2, "Initial learning rate.") flags.DEFINE_integer("num_neg_samples", 100, "Negative samples per training example.") flags.DEFINE_integer("batch_size", 16, "Number of training examples processed per step " "(size of a minibatch).") flags.DEFINE_integer("concurrent_steps", 12, "The number of concurrent training steps.") flags.DEFINE_integer("window_size", 5, "The number of words to predict to the left and right " "of the target word.") flags.DEFINE_integer("min_count", 5, "The minimum number of word occurrences for it to be " "included in the vocabulary.") flags.DEFINE_float("subsample", 1e-3, "Subsample threshold for word occurrence. Words that appear " "with higher frequency will be randomly down-sampled. Set " "to 0 to disable.") flags.DEFINE_boolean( "interactive", False, "If true, enters an IPython interactive session to play with the trained " "model. E.g., try model.analogy('france', 'paris', 'russia') and " "model.nearby(['proton', 'elephant', 'maxwell']") flags.DEFINE_integer("statistics_interval", 5, "Print statistics every n seconds.") flags.DEFINE_integer("summary_interval", 5, "Save training summary to file every n seconds (rounded " "up to statistics interval.") flags.DEFINE_integer("checkpoint_interval", 600, "Checkpoint the model (i.e. save the parameters) every n " "seconds (rounded up to statistics interval.") FLAGS = flags.FLAGS class Options(object): """Options used by our word2vec model.""" def __init__(self): # Model options. # Embedding dimension. self.emb_dim = FLAGS.embedding_size # Training options. # The training text file. self.train_data = FLAGS.train_data # Number of negative samples per example. self.num_samples = FLAGS.num_neg_samples # The initial learning rate. self.learning_rate = FLAGS.learning_rate # Number of epochs to train. After these many epochs, the learning # rate decays linearly to zero and the training stops. self.epochs_to_train = FLAGS.epochs_to_train # Concurrent training steps. self.concurrent_steps = FLAGS.concurrent_steps # Number of examples for one training step. self.batch_size = FLAGS.batch_size # The number of words to predict to the left and right of the target word. self.window_size = FLAGS.window_size # The minimum number of word occurrences for it to be included in the # vocabulary. self.min_count = FLAGS.min_count # Subsampling threshold for word occurrence. self.subsample = FLAGS.subsample # How often to print statistics. self.statistics_interval = FLAGS.statistics_interval # How often to write to the summary file (rounds up to the nearest # statistics_interval). self.summary_interval = FLAGS.summary_interval # How often to write checkpoints (rounds up to the nearest statistics # interval). self.checkpoint_interval = FLAGS.checkpoint_interval # Where to write out summaries. self.save_path = FLAGS.save_path # Eval options. # The text file for eval. self.eval_data = FLAGS.eval_data class Word2Vec(object): """Word2Vec model (Skipgram).""" def __init__(self, options, session): self._options = options self._session = session self._word2id = {} self._id2word = [] self.build_graph() self.build_eval_graph() self.save_vocab() self._read_analogies() def _read_analogies(self): """Reads through the analogy question file. Returns: questions: a [n, 4] numpy array containing the analogy question's word ids. questions_skipped: questions skipped due to unknown words. """ questions = [] questions_skipped = 0 with open(self._options.eval_data, "rb") as analogy_f: for line in analogy_f: if line.startswith(b":"): # Skip comments. continue words = line.strip().lower().split(b" ") ids = [self._word2id.get(w.strip()) for w in words] if None in ids or len(ids) != 4: questions_skipped += 1 else: questions.append(np.array(ids)) print("Eval analogy file: ", self._options.eval_data) print("Questions: ", len(questions)) print("Skipped: ", questions_skipped) self._analogy_questions = np.array(questions, dtype=np.int32) def forward(self, examples, labels): """Build the graph for the forward pass.""" opts = self._options # Declare all variables we need. # Embedding: [vocab_size, emb_dim] init_width = 0.5 / opts.emb_dim emb = tf.Variable( tf.random_uniform( [opts.vocab_size, opts.emb_dim], -init_width, init_width), name="emb") self._emb = emb # Softmax weight: [vocab_size, emb_dim]. Transposed. sm_w_t = tf.Variable( tf.zeros([opts.vocab_size, opts.emb_dim]), name="sm_w_t") # Softmax bias: [emb_dim]. sm_b = tf.Variable(tf.zeros([opts.vocab_size]), name="sm_b") # Global step: scalar, i.e., shape []. self.global_step = tf.Variable(0, name="global_step") # Nodes to compute the nce loss w/ candidate sampling. labels_matrix = tf.reshape( tf.cast(labels, dtype=tf.int64), [opts.batch_size, 1]) # Negative sampling. sampled_ids, _, _ = (tf.nn.fixed_unigram_candidate_sampler( true_classes=labels_matrix, num_true=1, num_sampled=opts.num_samples, unique=True, range_max=opts.vocab_size, distortion=0.75, unigrams=opts.vocab_counts.tolist())) # Embeddings for examples: [batch_size, emb_dim] example_emb = tf.nn.embedding_lookup(emb, examples) # Weights for labels: [batch_size, emb_dim] true_w = tf.nn.embedding_lookup(sm_w_t, labels) # Biases for labels: [batch_size, 1] true_b = tf.nn.embedding_lookup(sm_b, labels) # Weights for sampled ids: [num_sampled, emb_dim] sampled_w = tf.nn.embedding_lookup(sm_w_t, sampled_ids) # Biases for sampled ids: [num_sampled, 1] sampled_b = tf.nn.embedding_lookup(sm_b, sampled_ids) # True logits: [batch_size, 1] true_logits = tf.reduce_sum(tf.mul(example_emb, true_w), 1) + true_b # Sampled logits: [batch_size, num_sampled] # We replicate sampled noise lables for all examples in the batch # using the matmul. sampled_b_vec = tf.reshape(sampled_b, [opts.num_samples]) sampled_logits = tf.matmul(example_emb, sampled_w, transpose_b=True) + sampled_b_vec return true_logits, sampled_logits def nce_loss(self, true_logits, sampled_logits): """Build the graph for the NCE loss.""" # cross-entropy(logits, labels) opts = self._options true_xent = tf.nn.sigmoid_cross_entropy_with_logits( true_logits, tf.ones_like(true_logits)) sampled_xent = tf.nn.sigmoid_cross_entropy_with_logits( sampled_logits, tf.zeros_like(sampled_logits)) # NCE-loss is the sum of the true and noise (sampled words) # contributions, averaged over the batch. nce_loss_tensor = (tf.reduce_sum(true_xent) + tf.reduce_sum(sampled_xent)) / opts.batch_size return nce_loss_tensor def optimize(self, loss): """Build the graph to optimize the loss function.""" # Optimizer nodes. # Linear learning rate decay. opts = self._options words_to_train = float(opts.words_per_epoch * opts.epochs_to_train) lr = opts.learning_rate * tf.maximum( 0.0001, 1.0 - tf.cast(self._words, tf.float32) / words_to_train) self._lr = lr optimizer = tf.train.GradientDescentOptimizer(lr) train = optimizer.minimize(loss, global_step=self.global_step, gate_gradients=optimizer.GATE_NONE) self._train = train def build_eval_graph(self): """Build the eval graph.""" # Eval graph # Each analogy task is to predict the 4th word (d) given three # words: a, b, c. E.g., a=italy, b=rome, c=france, we should # predict d=paris. # The eval feeds three vectors of word ids for a, b, c, each of # which is of size N, where N is the number of analogies we want to # evaluate in one batch. analogy_a = tf.placeholder(dtype=tf.int32) # [N] analogy_b = tf.placeholder(dtype=tf.int32) # [N] analogy_c = tf.placeholder(dtype=tf.int32) # [N] # Normalized word embeddings of shape [vocab_size, emb_dim]. nemb = tf.nn.l2_normalize(self._emb, 1) # Each row of a_emb, b_emb, c_emb is a word's embedding vector. # They all have the shape [N, emb_dim] a_emb = tf.gather(nemb, analogy_a) # a's embs b_emb = tf.gather(nemb, analogy_b) # b's embs c_emb = tf.gather(nemb, analogy_c) # c's embs # We expect that d's embedding vectors on the unit hyper-sphere is # near: c_emb + (b_emb - a_emb), which has the shape [N, emb_dim]. target = c_emb + (b_emb - a_emb) # Compute cosine distance between each pair of target and vocab. # dist has shape [N, vocab_size]. dist = tf.matmul(target, nemb, transpose_b=True) # For each question (row in dist), find the top 4 words. _, pred_idx = tf.nn.top_k(dist, 4) # Nodes for computing neighbors for a given word according to # their cosine distance. nearby_word = tf.placeholder(dtype=tf.int32) # word id nearby_emb = tf.gather(nemb, nearby_word) nearby_dist = tf.matmul(nearby_emb, nemb, transpose_b=True) nearby_val, nearby_idx = tf.nn.top_k(nearby_dist, min(1000, self._options.vocab_size)) # Nodes in the construct graph which are used by training and # evaluation to run/feed/fetch. self._analogy_a = analogy_a self._analogy_b = analogy_b self._analogy_c = analogy_c self._analogy_pred_idx = pred_idx self._nearby_word = nearby_word self._nearby_val = nearby_val self._nearby_idx = nearby_idx def build_graph(self): """Build the graph for the full model.""" opts = self._options # The training data. A text file. (words, counts, words_per_epoch, self._epoch, self._words, examples, labels) = word2vec.skipgram(filename=opts.train_data, batch_size=opts.batch_size, window_size=opts.window_size, min_count=opts.min_count, subsample=opts.subsample) (opts.vocab_words, opts.vocab_counts, opts.words_per_epoch) = self._session.run([words, counts, words_per_epoch]) opts.vocab_size = len(opts.vocab_words) print("Data file: ", opts.train_data) print("Vocab size: ", opts.vocab_size - 1, " + UNK") print("Words per epoch: ", opts.words_per_epoch) self._examples = examples self._labels = labels self._id2word = opts.vocab_words for i, w in enumerate(self._id2word): self._word2id[w] = i true_logits, sampled_logits = self.forward(examples, labels) loss = self.nce_loss(true_logits, sampled_logits) tf.scalar_summary("NCE loss", loss) self._loss = loss self.optimize(loss) # Properly initialize all variables. tf.initialize_all_variables().run() self.saver = tf.train.Saver() def save_vocab(self): """Save the vocabulary to a file so the model can be reloaded.""" opts = self._options with open(os.path.join(opts.save_path, "vocab.txt"), "w") as f: for i in xrange(opts.vocab_size): f.write("%s %d\n" % (tf.compat.as_text(opts.vocab_words[i]), opts.vocab_counts[i])) def _train_thread_body(self): initial_epoch, = self._session.run([self._epoch]) while True: _, epoch = self._session.run([self._train, self._epoch]) if epoch != initial_epoch: break def train(self): """Train the model.""" opts = self._options initial_epoch, initial_words = self._session.run([self._epoch, self._words]) summary_op = tf.merge_all_summaries() summary_writer = tf.train.SummaryWriter(opts.save_path, graph_def=self._session.graph_def) workers = [] for _ in xrange(opts.concurrent_steps): t = threading.Thread(target=self._train_thread_body) t.start() workers.append(t) last_words, last_time, last_summary_time = initial_words, time.time(), 0 last_checkpoint_time = 0 while True: time.sleep(opts.statistics_interval) # Reports our progress once a while. (epoch, step, loss, words, lr) = self._session.run( [self._epoch, self.global_step, self._loss, self._words, self._lr]) now = time.time() last_words, last_time, rate = words, now, (words - last_words) / ( now - last_time) print("Epoch %4d Step %8d: lr = %5.3f loss = %6.2f words/sec = %8.0f\r" % (epoch, step, lr, loss, rate), end="") sys.stdout.flush() if now - last_summary_time > opts.summary_interval: summary_str = self._session.run(summary_op) summary_writer.add_summary(summary_str, step) last_summary_time = now if now - last_checkpoint_time > opts.checkpoint_interval: self.saver.save(self._session, opts.save_path + "model", global_step=step.astype(int)) last_checkpoint_time = now if epoch != initial_epoch: break for t in workers: t.join() return epoch def _predict(self, analogy): """Predict the top 4 answers for analogy questions.""" idx, = self._session.run([self._analogy_pred_idx], { self._analogy_a: analogy[:, 0], self._analogy_b: analogy[:, 1], self._analogy_c: analogy[:, 2] }) return idx def eval(self): """Evaluate analogy questions and reports accuracy.""" # How many questions we get right at precision@1. correct = 0 total = self._analogy_questions.shape[0] start = 0 while start < total: limit = start + 2500 sub = self._analogy_questions[start:limit, :] idx = self._predict(sub) start = limit for question in xrange(sub.shape[0]): for j in xrange(4): if idx[question, j] == sub[question, 3]: # Bingo! We predicted correctly. E.g., [italy, rome, france, paris]. correct += 1 break elif idx[question, j] in sub[question, :3]: # We need to skip words already in the question. continue else: # The correct label is not the precision@1 break print() print("Eval %4d/%d accuracy = %4.1f%%" % (correct, total, correct * 100.0 / total)) def analogy(self, w0, w1, w2): """Predict word w3 as in w0:w1 vs w2:w3.""" wid = np.array([[self._word2id.get(w, 0) for w in [w0, w1, w2]]]) idx = self._predict(wid) for c in [self._id2word[i] for i in idx[0, :]]: if c not in [w0, w1, w2]: return c return "unknown" def nearby(self, words, num=20): """Prints out nearby words given a list of words.""" ids = np.array([self._word2id.get(x, 0) for x in words]) vals, idx = self._session.run( [self._nearby_val, self._nearby_idx], {self._nearby_word: ids}) for i in xrange(len(words)): print("\n%s\n=====================================" % (words[i])) for (neighbor, distance) in zip(idx[i, :num], vals[i, :num]): print("%-20s %6.4f" % (self._id2word[neighbor], distance)) def _start_shell(local_ns=None): # An interactive shell is useful for debugging/development. import IPython user_ns = {} if local_ns: user_ns.update(local_ns) user_ns.update(globals()) IPython.start_ipython(argv=[], user_ns=user_ns) def main(_): """Train a word2vec model.""" if not FLAGS.train_data or not FLAGS.eval_data or not FLAGS.save_path: print("--train_data --eval_data and --save_path must be specified.") sys.exit(1) opts = Options() with tf.Graph().as_default(), tf.Session() as session: with tf.device("/cpu:0"): model = Word2Vec(opts, session) for _ in xrange(opts.epochs_to_train): model.train() # Process one epoch model.eval() # Eval analogies. # Perform a final save. model.saver.save(session, os.path.join(opts.save_path, "model.ckpt"), global_step=model.global_step) if FLAGS.interactive: # E.g., # [0]: model.analogy('france', 'paris', 'russia') # [1]: model.nearby(['proton', 'elephant', 'maxwell']) _start_shell(locals()) if __name__ == "__main__": tf.app.run()
