14 skills found
HaojiHu / Sets2SetsSequential sets to sequential sets learning
mathchi / Customer Segmentation With RFM AnalysisContext A real online retail transaction data set of two years. Content This Online Retail II data set contains all the transactions occurring for a UK-based and registered, non-store online retail between 01/12/2009 and 09/12/2011.The company mainly sells unique all-occasion gift-ware. Many customers of the company are wholesalers. Column Descriptors InvoiceNo: Invoice number. Nominal. A 6-digit integral number uniquely assigned to each transaction. If this code starts with the letter 'c', it indicates a cancellation. StockCode: Product (item) code. Nominal. A 5-digit integral number uniquely assigned to each distinct product. Description: Product (item) name. Nominal. Quantity: The quantities of each product (item) per transaction. Numeric. InvoiceDate: Invice date and time. Numeric. The day and time when a transaction was generated. UnitPrice: Unit price. Numeric. Product price per unit in sterling (£). CustomerID: Customer number. Nominal. A 5-digit integral number uniquely assigned to each customer. Country: Country name. Nominal. The name of the country where a customer resides. Acknowledgements Here you can find references about data set: https://archive.ics.uci.edu/ml/datasets/Online+Retail+II and Relevant Papers: Chen, D. Sain, S.L., and Guo, K. (2012), Data mining for the online retail industry: A case study of RFM model-based customer segmentation using data mining, Journal of Database Marketing and Customer Strategy Management, Vol. 19, No. 3, pp. 197-208. doi: [Web Link]. Chen, D., Guo, K. and Ubakanma, G. (2015), Predicting customer profitability over time based on RFM time series, International Journal of Business Forecasting and Marketing Intelligence, Vol. 2, No. 1, pp.1-18. doi: [Web Link]. Chen, D., Guo, K., and Li, Bo (2019), Predicting Customer Profitability Dynamically over Time: An Experimental Comparative Study, 24th Iberoamerican Congress on Pattern Recognition (CIARP 2019), Havana, Cuba, 28-31 Oct, 2019. Laha Ale, Ning Zhang, Huici Wu, Dajiang Chen, and Tao Han, Online Proactive Caching in Mobile Edge Computing Using Bidirectional Deep Recurrent Neural Network, IEEE Internet of Things Journal, Vol. 6, Issue 3, pp. 5520-5530, 2019. Rina Singh, Jeffrey A. Graves, Douglas A. Talbert, William Eberle, Prefix and Suffix Sequential Pattern Mining, Industrial Conference on Data Mining 2018: Advances in Data Mining. Applications and Theoretical Aspects, pp. 309-324. 2018. Inspiration This is Data Set Characteristics: Multivariate, Sequential, Time-Series, Text
ramzihaddad12 / Convex OptimizationGenerating collision-free trajectories in 3D space for multiple quadcopters within seconds by minimizing the total thrust at each time step for each quadcopter using Sequential Convex Programming (SCP). The goal is to transition from an initial to a final set of states, each consisting of position, velocity and acceleration. The vehicles must maintain a minimum distance between each other and satisfy other trajectory constraints.
HMY626 / MD4 CollisionMD4 Collisions MD4 is a 128-bit cryptographic hash function, meaning it should take a work factor of roughly 2^64 to find collisions. It turns out we can do much better. The paper "Cryptanalysis of the Hash Functions MD4 and RIPEMD" by Wang et al details a cryptanalytic attack that lets us find collisions in 2^8 or less. Given a message block M, Wang outlines a strategy for finding a sister message block M', differing only in a few bits, that will collide with it. Just so long as a short set of conditions holds true for M. What sort of conditions? Simple bitwise equalities within the intermediate hash function state, e.g. a[1][6] = b[0][6]. This should be read as: "the sixth bit (zero-indexed) of a[1] (i.e. the first update to 'a') should equal the sixth bit of b[0] (i.e. the initial value of 'b')". It turns out that a lot of these conditions are trivial to enforce. To see why, take a look at the first (of three) rounds in the MD4 compression function. In this round, we iterate over each word in the message block sequentially and mix it into the state. So we can make sure all our first-round conditions hold by doing this: # calculate the new value for a[1] in the normal fashion a[1] = (a[0] + f(b[0], c[0], d[0]) + m[0]).lrot(3) # correct the erroneous bit a[1] ^= ((a[1][6] ^ b[0][6]) << 6) # use algebra to correct the first message block m[0] = a[1].rrot(3) - a[0] - f(b[0], c[0], d[0]) Simply ensuring all the first round conditions puts us well within the range to generate collisions, but we can do better by correcting some additional conditions in the second round. This is a bit trickier, as we need to take care not to stomp on any of the first-round conditions. Once you've adequately massaged M, you can simply generate M' by flipping a few bits and test for a collision. A collision is not guaranteed as we didn't ensure every condition. But hopefully we got enough that we can find a suitable (M, M') pair without too much effort. Implement Wang's attack.
rahafaljundi / Expert GateWe introduce a model of lifelong learning, based on a Network of Experts. New tasks / experts are learned and added to the model sequentially, building on what was learned before. To ensure scalability of this process, data from previous tasks cannot be stored and hence is not available when learning a new task. A critical issue in such context, not addressed in the literature so far, relates to the decision of which expert to deploy at test time. We introduce a set of gating autoencoders that learn a representation for the task at hand, and, at test time, automatically forward the test sample to the relevant expert. This also brings memory efficiency as only one expert network has to be loaded into memory at any given time. Further, the autoencoders inherently capture the relatedness of one task to another, based on which the most relevant prior model to be used for training a new expert, with finetuning or learning without-forgetting, can be selected. We evaluate our method on image classification and video prediction problems.
malikfahad / Particle FilterParticle filters or Sequential Monte Carlo (SMC) methods are a set of genetic, Monte Carlo algorithms used to solve filtering problems arising in signal processing and Bayesian statistical inference. The filtering problem consists of estimating the internal states in dynamical systems when partial observations are made, and random perturbations are present in the sensors as well as in the dynamical system. The objective is to compute the posterior distributions of the states of some Markov process, given some noisy and partial observations. Particle filters implement the prediction-updating transitions of the filtering equation directly by using a genetic type mutation-selection particle algorithm. The samples from the distribution are represented by a set of particles; each particle has a likelihood weight assigned to it that represents the probability of that particle being sampled from the probability density function.
drhema / Evolution Api InstallationThis guide outlines the necessary steps to install MongoDB, Redis, Nginx, and the Evolution API. Follow these steps sequentially to set up your environment properly.
fazildgr8 / AppliedDeepLearningThis repository consists a set of Jupyter Notebooks with a different Deep Learning methods applied. Each notebook gives walkthrough from scratch to the end results visualization hierarchically. The Deep Learning methods include Multiperceptron layers, CNN, GAN, Autoencoders, Sequential and Non-Sequential deep learning models. The fields applied includes Image Classification, Time Series Prediction, Recommendation Systems , Anomaly Detection and Data Analysis.
anunay2608 / DLVSA novel deep learning based visual servoing architecture “DLVS” is proposed for control of an unmanned aerial vehicle (UAV) capable of quasi-stationary flight with a camera mounted under the vehicle to track a target consisting of a finite set of stationary points lying in a plane. Current Deep Learning and Reinforcement Learning (RL) based end-to-end servoing approaches rely on training convolutional neural networks using color images with known camera poses to learn the visual features in the environment suitable for servoing tasks. This approach limits the application of the network to known environments where the dataset was collected, moreover such networks cannot be deployed on the low power computers present on board the UAV. The proposed solution employs a time series architecture to learn temporal data from sequential values to output the control ques to the flight controller. The low computational complexity and flexibility of the DLVS architecture ensures real time on board tracking for virtually any target. The algorithm was thoroughly validated in real-life environments and outperformed the current state-of-the-arts in terms of both time efficiency and accuracy.
hubertdungen / MonitorSwitcherA simple, lightweight PowerShell script designed to switch your primary display between multiple monitors on a Windows-based system. The script keeps a log to remember which monitor was last set as the primary one and alternates sequentially through all available monitors.
Saannndddyyyyy / Face Recognition System Using Parallel ComputingThe face recognizing systems available so far maintain a database that has each pre-processed human face and their corresponding unique features as determined from each face. These features are stored with the respective individual’s face. Therefore, when a query is raised, the unique features are extracted from the face and then compared with the features in the database and results in a perfect match .There are many areas of applications that use face recognition technology ranging from security applications for recognizing criminals in public spaces such as airports, and shopping centers, verifying access to private property, and casting votes, to intelligent vision-based human computer interaction such as ATM, cell phones and intelligent buildings. Identifying the images and processing the data is a challenging task because of the various factors involved in this sophisticated process such as illumination, angle of pose, accessories, facial expression, and aging effects. There are two sets of data involved in the face recognition system. The first is the training set of data that is used in the learning stage. The second is a testing set which is used during recognition. Often real time response is understood to be in order of milliseconds and sometimes microseconds, which is the most crucial criteria in the system design. All in all, it can be said that this project is a good example of a parallel task with omp. Dealing with multiple pieces of data to read, calculate, and compare results takes a long time when tasks have to be performed sequentially.
ayushc13 / 32 Bit RISC Processor Using HDL VerilogThe proposed processor is designed using HDL Verilog having separate instruction and data memory. The salient feature of proposed processor is pipelining, used for improving performance, such that on every clock cycle one instruction will be executed. Another important feature is that instruction set contains 14 instructions, which is very simple, easy to learn and compact. The proposed processor has 32-bit ALU, Thirty two 32-bit general-purpose registers, no flag register and memory word size of 32 bits. Another advantage of the proposed processor is that it can execute programs with as many instructions with very few addressing modes like register, immediate, register indexed etc. such that any practical programs can be fitted into it. The proposed processor is physically verified on Xilinx Spartan 6 FPGA board having chipset XC6SLX9 with 50MHz clock cycle. HDL Verilog is quite different from Implementation in sequential languages mainly because of the parallel nature of the HDLs, where sequential language code is executed by a digital core step by step, HDL code describes the functioning of a digital hardware, it is taken by synthesis tools that try to find a digital hardware implementation of the description, thus there is no step by step execution of the statement of HDL, and each statement is a smaller circuit in itself.
carloselcastro / MandelbrotThe main goal of this work is to generate the Mandelbrot Set using different approaches: Sequential, Pthreads, OpenMP, OpenMPI and Cuda.
artlessconstruct / Cloudflare Worker Email ForwarderA Cloudflare Email Worker providing configurable email forwarding which routes from email addresses using sub-addressing (a.k.a. RFC 5233 sub-address extension, tagged addresses, plus addresses, etc.) to a set of destinations simultaneously, where each such destination is a sequence of addresses attempted sequentially until one succeeds.
