GitLab

An earlier change made it so running this file would cause an error. The unit tests still passed, as the unit tests do not directly call the main() function

PiperOrigin-RevId: 264268400

* 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