tf.mlir.experimental.convert_function

Import a ConcreteFunction and convert it to a textual MLIR module.

tf.mlir.experimental.convert_function(
    concrete_function,
    pass_pipeline='tf-standard-pipeline',
    show_debug_info=False
)

This API is only intended for inspecting the internals of TensorFlow and the string returned is at the moment intended for debugging purposes.

A tf.function can be imported and converted from TensorFlow to TensorFlow MLIR with this API by extracting its ConcreteFunction (eagerly-executing wrapper around a tf.Graph).

For example:

@tf.function
def add(a, b):
  return a + b
 
concrete_function = add.get_concrete_function(
    tf.TensorSpec(None, tf.dtypes.float32),
    tf.TensorSpec(None, tf.dtypes.float32))
tf.mlir.experimental.convert_function(concrete_function)
'...module attributes {...} {...}...'

concrete_function An object of type ConcreteFunction.
pass_pipeline A textual description of an MLIR Pass Pipeline to run on the module, see MLIR documentation for the textual pass pipeline syntax.
show_debug_info Whether to include locations in the emitted textual form.

A textual representation of the MLIR module corresponding to the ConcreteFunction.

InvalidArgumentError if concrete_function is invalid or cannot be converted to MLIR.