MLFinalProject
ML for Cyber Sec; CSAW-HackML-2020
Authors:
Tieliu Zhou, tz1430, N14973953;
Juntao Jiang, jj2910, N10249972;
Kangning Li, kl3596, N12602103;
Jiazhen Han, jh6419, N19591264;
Repo Architecture:
Models: fixed models;
Src: the raw code and help scripts;
How to Run Code:
For approach1(without PCA src/Fine-Pruning Approach): There is only a colab file with code and description inside.
For approach2(use PCA src/Fine-Pruning-PCA-Approach): There are train.sh and infer.sh to do the training part and inference part. Please make sure the data is in the right directory(MLFinalProject/src/Fine-Pruning-PCA-Approach/data).
For single image: Please use the eval script and put the test image into the data floder(MLFinalProject/data).
For evaluating the model, execute eval.py by running:
python3 eval_*.py </data directory>
E.g
python3 eval_anonymous_1.py data/test_image.png