GitHub

tfm.nlp.serving_modules.features_to_int32

Converts tf.int64 features to tf.int32, keep other features the same.

tfm.nlp.serving_modules.features_to_int32(
    features: Dict[str, tf.Tensor]
) -> Dict[str, tf.Tensor]

tf.Example only supports tf.int64, but the TPU only supports tf.int32.

Args
`features`	Input tensor dictionary.

Returns
Features with tf.int64 converted to tf.int32.

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates. Some content is licensed under the numpy license.

Last updated 2023-01-20 UTC.

Args