While working on Images in Machine Learning Projects it is very important to Pre-process the images. Preprocessing the images takes a lot of time and is very error-prone process
we have built and developed the pre-processing Library that pre-process the images, in this we have :
- read N no of images from N no of directories
- read the classes of every image
- Convert them to Numpy array
- Normalize-data
- One-hot-encode-data
- reshape-data [conversion in required dimension]
Just you need to provide the Image directory and the reshape size
Eg :
training_path = '/Kaggle/training_set/' testing_path = '/Kaggle/test_set/'
train_obj = DataPreprocessing(training_path, 300) train_images, train_labels = train_obj.load_data()
test_obj = DataPreprocessing(testing_path, 300) test_images, test_labels = test_obj.load_data()