GitLab

* Merged commit includes the following changes:
263863588  by yongzhe:

    Fix a bug that the SetExternalContext for EdgeTPU wasn't called when initializing LSTD client.

--
263370193  by yongzhe:

    Internal change.

--

PiperOrigin-RevId: 263863588

* Revert changes in seq_dataset_builder_test.py

PiperOrigin-RevId: 264244022

PiperOrigin-RevId: 264196955

PiperOrigin-RevId: 264193390

PiperOrigin-RevId: 264174941

PiperOrigin-RevId: 264174913

PiperOrigin-RevId: 263910123

PiperOrigin-RevId: 263879950

PiperOrigin-RevId: 263874363

PiperOrigin-RevId: 263863438

PiperOrigin-RevId: 263858817

PiperOrigin-RevId: 263854996

PiperOrigin-RevId: 263837214

Also add `worker_hosts` and `task_index` flags.  These flags enable running the
model over multiple hosts by passing the cluster information via command line.

Setting `TF_CONFIG` will continue to work.

PiperOrigin-RevId: 263825245

PiperOrigin-RevId: 263812316

PiperOrigin-RevId: 263669433

PiperOrigin-RevId: 263485329

PiperOrigin-RevId: 263463300

PiperOrigin-RevId: 263448204

PiperOrigin-RevId: 263401952

PiperOrigin-RevId: 263261200

PiperOrigin-RevId: 263257133

libedgetpu.h was used instead of edgetpu.h:
https://coral.googlesource.com/edgetpu-native/+/refs/heads/release-diploria/libedgetpu/

PiperOrigin-RevId: 263176146

PiperOrigin-RevId: 263204353

PiperOrigin-RevId: 263257133

PiperOrigin-RevId: 263255352

PiperOrigin-RevId: 263230689

PiperOrigin-RevId: 263217684

PiperOrigin-RevId: 263158478

263011341  by hongkuny<hongkuny@google.com>:

    reduce batch size for bert squad v8 accuracy test

--

PiperOrigin-RevId: 263011341

* Replace google3.pyglib modules with tf and absl

This now matches train.py and provides more publicly available libraries.

* Add example pipeline config for SSD Interleaved V2 Model.

Compiled from model_builder_test.py and lstm_ssd_mobilenet_v1_imagenet.config,
Removed data augmentation and tranfer learning (i.e. training from checkpoint) due to errors I was seeing when trying to run with it.

* Remove unused tfrecord creation.

This was also incorrectly specified, as the keys differed from the TFSequenceExample parser.

* correct key specified in docstring

* add tflite frozen graph exporter (cli and lib).

* add tflite model exporter

* add script to test the tflite model

* add mode export documentation

* correct docstring

* rename export files to be unique across detection research work

* correct number of channels for grayscale

* add and correct copyright

262988559  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Enable NCF TF 2.0 model to run on TPUStrategy.

--
262971756  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

262967691  by hongkuny<hongkuny@google.com>:

    Internal

--

PiperOrigin-RevId: 262988559