API - Visualization

TensorFlow provides TensorBoard to visualize the model, activations etc. Here we provide more functions for data visualization.

tensorlayer.visualize

read_image read_images save_image save_images draw_boxes_and_labels_to_image draw_mpii_pose_to_image draw_weights CNN2d frame images2d tsne_embedding

Save and read images

Read one image

read_image

Read multiple images

read_images

Save one image

save_image

Save multiple images

save_images

Save image for object detection

draw_boxes_and_labels_to_image

Save image for pose estimation (MPII)

draw_mpii_pose_to_image

Visualize model parameters

Visualize CNN 2d filter

CNN2d

Visualize weights

draw_weights

Visualize images

Image by matplotlib

frame

Images by matplotlib

images2d

Visualize embeddings

tsne_embedding