hi bois

First, get the images in their carrying, normal, threat folders Then run the following:

# sorts them into training, validation, and testing sets
$ python3 data_classifier.py

# creates the datasets that pytorch uses
$ python3 dataset.py

# trains the model
$ python3 main.py

# evaluates the model against the test set, where X is the model number
$ python3 inference.py --m ./model_X

# compares your inference with the actual test set
$ python3 compare.py

If you want to test your model on one image or on some other directory, can do

$ python3 inference.py --m <path to model> --t <path to image>