Predicting Invasive Ductal Carcinoma using Convolutional Neural Network (CNN) in Keras

The dataset consists of 279 folders, with sub-folders 0 and 1 inside each of the 279 folders. We then create a function process_images which takes as input the starting and end index of the images. This function first reads the image using OpenCV's cv2.imread() and also resizes the image. Resizing is done because few of the images in the dataset are not 50x50x3. The function returns two arrays: X, which is an array of the resized image data and Y, which is an array of the corresponding labels.