tf.lite.experimental.authoring.compatible
Wraps tf.function
into a callable function with TFLite compatibility checking.
tf.lite.experimental.authoring.compatible( target=None, converter_target_spec=None, **kwargs )
Example:
@tf.lite.experimental.authoring.compatible @tf.function(input_signature=[ tf.TensorSpec(shape=[None], dtype=tf.float32) ]) def f(x): return tf.cosh(x) result = f(tf.constant([0.0])) # COMPATIBILITY WARNING: op 'tf.Cosh' require(s) "Select TF Ops" for model # conversion for TensorFlow Lite. # Op: tf.Cosh # - tensorflow/python/framework/op_def_library.py:748 # - tensorflow/python/ops/gen_math_ops.py:2458 # - <stdin>:6
Args | |
---|---|
target |
A tf.function to decorate. |
converter_target_spec |
target_spec of TFLite converter parameter. |
**kwargs |
The keyword arguments of the decorator class _Compatible. |
Returns | |
---|---|
A callable object of tf.lite.experimental.authoring._Compatible . |
© 2022 The TensorFlow Authors. All rights reserved.
Licensed under the Creative Commons Attribution License 4.0.
Code samples licensed under the Apache 2.0 License.
https://www.tensorflow.org/versions/r2.9/api_docs/python/tf/lite/experimental/authoring/compatible