/COS495-project

Classification for images of driving.

Primary LanguagePython

COS495-project

To run network, create two files that contain (a) training images and (b) validation images, in a .txt of the form

filepathToImage label

Then run

python statefarm_train.py <train.txt> <validate.txt>

You must ensure the dimensions of your training and validation images match each other and the image dimensions on line 30 of statefarm_input.py

IMAGE_SIZE = (75, 100)

To make predictions on a set of (unlabeled) testing data, create a .txt containing one filepath per line. Then run

python statefarm_eval.py <test.txt>