datasets/raise/split_and_compare_prnus_of_subgroups.py

@@ -5,8 +5,6 @@ from tqdm import tqdm
 from utils import denoise
 from skimage import img_as_float
 import sys
-import os
-from random import shuffle
 
 sys.path.insert(0, '../../algorithms/distance/')
 
@@ -14,19 +12,19 @@ from rms_diff import rmsDiffNumpy
 
 NUMBER_OF_SUBGROUPS = 1
 DENOISER = 'wavelet'
-IMAGES_FOLDER = 'flat-field/TIF'
 
+IMAGES_FOLDER = 'flat-field/TIF'
 imagesFileNames = os.listdir(IMAGES_FOLDER)
-# To not have a bias (chronological for instance) when split to make subgroups.
-shuffle(imagesFileNames)
 numberOfImagesPerSubgroup = len(imagesFileNames) // NUMBER_OF_SUBGROUPS
+
+# 1
 numberOfImagesThresholds = range(numberOfImagesPerSubgroup, numberOfImagesPerSubgroup + 1)
 
+# Assume random image order to not introduce a bias.
 subgroupsPrnuEstimatesNpArray = []
-for subgroupIndex in tqdm(range(NUMBER_OF_SUBGROUPS), 'Subgroup'):
+for subgroupIndex in range(NUMBER_OF_SUBGROUPS):
     imagesPrnuEstimateNpArray = []
-    subgroupImagesFileNames = imagesFileNames[numberOfImagesPerSubgroup * subgroupIndex : numberOfImagesPerSubgroup * (subgroupIndex + 1)]
-    for imageFileName in tqdm(subgroupImagesFileNames, f'Image of subgroup {subgroupIndex}'):
+    for imageFileName in tqdm(imagesFileNames[numberOfImagesPerSubgroup * subgroupIndex : numberOfImagesPerSubgroup * (subgroupIndex + 1)]):
         imagePath = f'{IMAGES_FOLDER}/{imageFileName}'
         imagePil = Image.open(imagePath)
         imageNpArray = img_as_float(np.array(imagePil))
@@ -41,7 +39,7 @@ for subgroupIndex in tqdm(range(NUMBER_OF_SUBGROUPS), 'Subgroup'):
     subgroupsPrnuEstimatesNpArray += [subgroupPrnuEstimateNpArray]
 
 rmss = []
-for numberOfImagesIndex in range(len(numberOfImagesThresholds)):
+for numberOfImagesIndex, numberOfImages in enumerate(numberOfImagesThresholds):
     rms = rmsDiffNumpy(subgroupsPrnuEstimatesNpArray[0][numberOfImagesIndex], subgroupsPrnuEstimatesNpArray[1][numberOfImagesIndex])
     rmss += [rms]