Clean-up

README.md

@@ -71,6 +71,16 @@ digest = hashing.total_hashing(algos, thresholds, path_database, positive_datase
 utils.save_digest(digest, save_folder)
 ```
 
-All this is contained in `main.py`.
+All this is contained in `main.py`. 
 
-The final digest is composed of 6 files : `general.json` with general metrics for all the experiment, `attacks.json` containing the metrics for each types of attack, `images_pos.json` and `images_neg.json` containing number of correct/incorrect detection for each image in the database respectively, and `match_time.json` and `db_time.json` respectively containing the time (s) for the matching phase and the the database creation phase.
+The final digest is composed of 6 files : `general.json` with general metrics for all the experiment, `attacks.json` containing the metrics for each types of attack, `images_pos.json` and `images_neg.json` containing number of correct/incorrect detection for each image in the database respectively, and `match_time.json` and `db_time.json` respectively containing the time (s) for the matching phase and the the database creation phase.
+
+To process and create figures from the digest, one can look into `process.py`.
+
+# Datasets
+
+We personally used 3 datasets that can be found at <INSERT LINK !!!> and for which the splitting has already been done. They are the BSDS500 dataset, ImageNet validation set and a dataset from Kaggle containing memes from reddit.
+
+# Pre-trained SimCLR models 
+
+The pre-trained SimCLR models are not available in this repository due to their large size. They can be downloaded (along with their architecture definition) directly in the [github of the authors ](https://github.com/google-research/simclr) or downloaded directly at <INSERT LINK !!!> (folders SimCLRv1 and SimCLRv2) and then added to the `hashing` folder of the library.

process.py

@@ -164,60 +164,5 @@
     res[algo] = {'fpr': fpr[index], 'threshold': thresholds[index]}
 
 
-#%%
-import os
-from helpers import utils
-from helpers import create_plot as plot
-
-experiment_names = [f'Database_{i}_ImageNet' for i in [250, 2500, 25000]]
-experiment_folders = ['Results/' + a for a in experiment_names]
-
-res = []
-
-for a in experiment_folders:
-
-    general, _, _, _, _,_ = utils.load_digest(a)
-    res.append(general)
-
-plot.roc_comparison_database(res, save=False, filename='Comparison_db_size.pdf')
-
-#%%
-from helpers import utils
-
-experiment_folder = 'Results/Test_memes1/'
-
-general, image_pos, image_neg, _, _ = utils.load_digest(experiment_folder, False)
-
-
-#%%
-
-from PIL import Image
-import cv2
-import numpy as np
-
-original = 'Datasets/BSDS500/Control/data17.jpg'
-original = Image.open(original)
-original = np.array(original.convert('L'))
-
-copy = 'Datasets/BSDS500/Control_attacks/data17_rotation_60_and_rescaling.png'
-copy = Image.open(copy)
-copy = np.array(copy.convert('L'))
-
-sift = cv2.SIFT_create()
-
-kp1, des1 = sift.detectAndCompute(original, None)
-kp2, des2 = sift.detectAndCompute(copy, None)
-
-bf = cv2.BFMatcher(cv2.NORM_L2, crossCheck=True)
-matches = bf.match(des1,des2)
-matches = sorted(matches, key = lambda x:x.distance)
-
-img3 = cv2.drawMatches(original,kp1,copy,kp2,matches[:30],None,flags=cv2.DrawMatchesFlags_NOT_DRAW_SINGLE_POINTS)
-
-out = Image.fromarray(img3)
-# out.save('test_feat.png')
 
-img = cv2.drawKeypoints(original, np.random.choice(kp1, 500, replace=False), None, flags=cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)
-out_keypoints = Image.fromarray(img)
-out_keypoints.save('keypoints.png')