Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- import numpy as np
- # Number of components in each image (N)
- num_components = 100
- # Number of images in the training set (K)
- num_images = 1000
- # Number of classes (M)
- num_classes = 10
- # Create an empty training set
- X = []
- # Create an empty list for class labels
- Y = []
- for i in range(num_classes):
- # Generate random sample images for each class
- class_images = np.random.rand(num_images, num_components)
- # Add the images to the training set
- X.append(class_images)
- # Generate class labels for each image
- class_labels = np.full(num_images, i)
- # Add the labels to the class labels list
- Y.append(class_labels)
- # Concatenate the images and labels of all classes
- X = np.concatenate(X)
- Y = np.concatenate(Y)
- # Generate reference images (etalons) for each class
- etalons = np.random.rand(num_classes, num_components)
Advertisement
Add Comment
Please, Sign In to add comment