tfp.experimental.auto_batching.stackless.is_running

View source on GitHub

Returns whether the stackless machine is running a computation.

Aliases:

tfp.experimental.auto_batching.stackless.is_running()

This can be useful for writing special primitives that change their behavior depending on whether they are being staged, run stackless, inferred (see type_inference.is_inferring), or none of the above (i.e., dry-run execution, see frontend.Context.batch).

Returns:

  • running: Python bool, True if this is called in the dynamic scope of stackless running, otherwise False.