PHDR

<p align="center"><img width=40% src="https://github.com/ralireza/phdr/blob/master/media/logo.png"></p>

                                          

Classifier

• knn
• parzen-window
• bayes
• mlp
• rbf
• svm
• random forest
• decision tree

Feature Extractor

• PCA
• HOG
• RESIZE

Report

Algorithm | Feature-Selector | Accuracy | --- | --- | --- KNN | PCA | 0.9926 KNN | HOG | 0.9889 KNN | RESIZE | 0.9933 parzen-window | PCA | 0.9852 parzen-window | HOG | 0.9927 parzen-window | RESIZE | 1 bayes | PCA | 0.9114 bayes | HOG | 0.8923 bayes | RESIZE | 0.9704 mlp | PCA | 0.9926 mlp | HOG | 0.9963 mlp | RESIZE | 0.20 rbf | PCA | 0.9926 rbf | HOG | 0.9926 rbf | RESIZE | 0.9852 svm | PCA | 0.9926 svm | HOG | 0.9852 svm | RESIZE | 0.9926 random forest | PCA | 0.9926 random forest | HOG | 0.9963 random forest | RESIZE | 0.9963 decision tree | PCA | 0.9889 decision tree | HOG | 1 decision tree | RESIZE | 0.9926

Usage

python3 PHDR.py -c [classifier] -f [feature-selector]

Help

 -h, --help            show this help message and exit
  -c CLASSIFIER, --classifier CLASSIFIER
                        knn parzen bayes mlp rbf rforest dtree svm
  -f FEATUESELECT, --featueselect FEATUESELECT
                        pca hog resize
  -v, --version         show program's version number and exit