Completed
t.print: may change runtime graph optimization
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Inside TensorFlow - TF Debugging
Automatically move to the next video in the Classroom when playback concludes
- 1 Intro
- 2 Scope of this talk: "debugging" is an overloaded term in ML
- 3 Printing Eager Tensor values
- 4 Printing the value of graph-internal tensors
- 5 Homework: tf.print() on composite tensors
- 6 Programmatically access graph-internal tensor values
- 7 Programmatically fetching graph-internal tensors: While loop?
- 8 Finding device placement: Pure eager execution
- 9 Finding out device placement: tf.function
- 10 Getting and plotting the graph of a function: Colab (google3 only)
- 11 Dumping Grappler outputs: The graph that actually (almost) gets executed at runtime (bazel builds)
- 12 t.print: may change runtime graph optimization
- 13 t.config.experimental_run_functions_eagerly
- 14 Step debugging: Using tf.config.experimental_run_functions_eagerly
- 15 Step debugging: What happens inside a non-eagerly-executing function?
- 16 tf.config.experimental_run_functions eagerly does not work on tf.data.Dataset.mapo
- 17 Getting Access to tf.keras Layer Activations
- 18 Debugging Keras Models with TensorBoard callback
- 19 Parting notes
