CONTENTS:

The folder contains 7 files and 2 folders:

1. Files

1.1. constants.py <br>
1.2. data_prep.py <br>
1.3. prediction.py <br>
1.4. roles.json <br>
1.5. telegramnotification3.py <br>
1.6. trained_knn_model.clf <br>
1.7. training.py <br>
1.8. README.txt <br>

2. Folders
   2.1. image dataset: The folders gets updated dynamically with the code
   2.2 analysis: Contains the required dataset and code used for face encoding and encoding classification analysis
   

Required Packages:

• cv2
  
• skimage
  
• face_recognition
  
• telegram-bot
  
• PIL
  
• seaborn, keras (These packages are only required for analysis. Main code can run without it)
  

How to run the code:

* Smart Visitor Recognition code:

• Make sure you are in the source directory
  
• Run prediction.py file to predict the person who comes to the door.
  
• Run data_prep.py to register a new person to the system.
  

* Analysis Code:

• Run knn_analysis.ipynb notebook to run the code for KNN classifier.
  
• Run svm_analysis.ipynb notebook to run the analysis for the SVM classifier.
  
• Run encoding_analysis.ipynb notebook the get analysis on encoding method comparison.
  
• Train_test.py code is to be run to make a separate testing folder. This code only needs to run one time and has already been executed so no need to run again.