Skip to content
This repository was archived by the owner on Jun 9, 2021. It is now read-only.
This repository was archived by the owner on Jun 9, 2021. It is now read-only.

op type not registered NormalizeUTF8 initializing BERT #276

Open
@jaismith

Description

@jaismith

Getting the following error when initializing a BERT model using this package:

FileNotFoundError: Op type not registered 'NormalizeUTF8' in binary running on <computer name>.local. Make sure the Op and Kernel are registered in the binary running in this process. Note that if you are loading a saved graph which used ops from tf.contrib, accessing (e.g.) `tf.contrib.resampler` should be done before importing the graph, as contrib ops are lazily registered when the module is first accessed.
 If trying to load on a different device from the computational device, consider using setting the `experimental_io_device` option on tf.saved_model.LoadOptions to the io_device such as '/job:localhost'.

Here's the code initializing the model:

# use ALBERT
tfhub_handle_encoder = 'https://tfhub.dev/tensorflow/albert_en_base/2'
tfhub_handle_preprocess = 'https://tfhub.dev/tensorflow/albert_en_preprocess/3'

bert_preprocess_model = hub.KerasLayer(tfhub_handle_preprocess)
bert_model = hub.KerasLayer(tfhub_handle_encoder)

I've tried installing via the quick install script and the Conda instructions in #153, both throw the same error. Would appreciate if anyone had any insights.


Edit: modules are imported as follows:

import tensorflow as tf
import tensorflow_hub as hub
import tensorflow_addons.text as text
import tensorflow_addons.optimizers as optimizers

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions