March 2019
Intermediate to advanced
196 pages
4h 50m
English
TensorBoard is a visualization tool for trained models. The full signature of the TensorBoard callback is as follows:
tf.keras.callbacks.TensorBoard(log_dir='./logs', histogram_freq=0, batch_size=32, write_graph=True, write_grads=False, write_images=False, embeddings_freq=0, embeddings_layer_names=None, embeddings_metadata=None, embeddings_data=None, update_freq='epoch')
There is a very clear and detailed description of all of these arguments at https://keras.io/callbacks/. TensorBoard may be invoked from the command line like so:
tensorboard --logdir=/full_path_to_your_logs
For example, we can use tensorboard --logdir=./logs for the default directory. Setting histogram_freq to anything other than 0 causes a significant ...
Read now
Unlock full access