GitLab

PiperOrigin-RevId: 314198383

PiperOrigin-RevId: 314177321

Move common configs inside base_config to config_definitions.

PiperOrigin-RevId: 314074592

PiperOrigin-RevId: 314002442

PiperOrigin-RevId: 313952251

PiperOrigin-RevId: 313906815

PiperOrigin-RevId: 313867862

PiperOrigin-RevId: 313865079

Moved code from TF1 to TF2 for DELF logging, testing directories and feature extraction scripts (#8591)

* Merged commit includes the following changes:

FolderOrigin-RevId: /google/src/cloud/bfusinska/delf

* Added the import for utils
Co-authored-by: Andre Araujo <andrearaujo@google.com>

Proposes the full functionality of MultiHeadAttention layer. This change first goes to model garden NLP library.

PiperOrigin-RevId: 313847485

PiperOrigin-RevId: 313812017

PiperOrigin-RevId: 313806578

PiperOrigin-RevId: 313798102

PiperOrigin-RevId: 313794704

PiperOrigin-RevId: 313711327

PiperOrigin-RevId: 313708781

PiperOrigin-RevId: 313693539

PiperOrigin-RevId: 313662797

Deprecate old customized training loop for run_classifier.py as compile/fit fully satisfy needs/performance.

PiperOrigin-RevId: 313660745

Float32 is used if the model uses mixed precision with bfloat16. Float16 activation are unchanged.

The motivation is that BERT with the LAMB optimizer with a gelu activation has an unstable loss when gelu is in bfloat16. Unfortunately, it is not easy to check if the LAMB optimizer and gelu is used, and perhaps there are other cases that work better with float32 activations instead of bfloat16 activations, so we always do the activation in float32 instead of bfloat16.

PiperOrigin-RevId: 313618322

PiperOrigin-RevId: 313536026

PiperOrigin-RevId: 313475975

PiperOrigin-RevId: 313435294

PiperOrigin-RevId: 313321531

PiperOrigin-RevId: 313313579

* Merged commit includes the following changes:
311933687  by Sergio Guadarrama:

    Removes spurios use of tf.compat.v2, which results in spurious tf.compat.v1.compat.v2. Adds basic test to nasnet_utils.
    Replaces all remaining import tensorflow as tf with import tensorflow.compat.v1 as tf

--
311766063  by Sergio Guadarrama:

    Removes explicit tf.compat.v1 in all call sites (we already import tf.compat.v1, so this code was  doing tf.compat.v1.compat.v1). The existing code worked in latest version of tensorflow, 2.2, (and 1.15) but not in 1.14 or in 2.0.0a, this CL fixes it.

--
311624958  by Sergio Guadarrama:

    Updates README that doesn't render properly in github documentation

--
310980959  by Sergio Guadarrama:

    Moves research_models/slim off tf.contrib.slim/layers/framework to tf_slim

--
310263156  by Sergio Guadarrama:

    Adds model breakdown for MobilenetV3

--
308640...

PiperOrigin-RevId: 313259032

PiperOrigin-RevId: 313205490

Add a link for DELG

PiperOrigin-RevId: 313148142

This allows one to finetune a BERT model into a task before using it for
another task. E.g. SQuAD before finetune another QA type of tasks.

PiperOrigin-RevId: 313145768

PiperOrigin-RevId: 313125068

PiperOrigin-RevId: 313032520

PiperOrigin-RevId: 313030129

PiperOrigin-RevId: 312988874

PiperOrigin-RevId: 312945782

PiperOrigin-RevId: 312939899

PiperOrigin-RevId: 312923051

PiperOrigin-RevId: 312889153

PiperOrigin-RevId: 312841381