GitLab

Add Overfit and Underfit notebook.

LFADS: Support for posterior_push_mean instead of sampling

Rename eager get started notebook to custom training walkthrough

Add MIT license

* remove unused imports and lint

* fix schedule.py

* address PR comments

PiperOrigin-RevId: 200493322

Suppressed FutureWarning in create_pet_tf_record

Added new model, global objectives.

Fixing bug for evaluating the quantized model in research/object_detection

Attempt to hide copyright notice. Remove some italics.

Fix the Bug that evaluator don't load the min max values of fake_quantize op from lastest checkpoint

Fixed variable name typo for deeplab_demo Tutorial

Clarified .bashrc Configuration in G3Doc Installation

Minor variable name fix

* Add checklist for official models. Remove file access from flag validator (causing issues with BUILD)

* spelling

* address PR comments

* Add DistributionStrategy to transformer model

* add num_gpu flag

* Calculate per device batch size for transformer

* remove reference to flags_core

* Add synthetic data option to transformer

* fix typo

* add import back in

* Use hierarchical copy

* address PR comments

* lint

* fix spaces

* group train op together to fix single GPU error

* Fix translate bug (sorted_keys is a dict, not a list)

* Change params to a default dict (translate.py was throwing errors because params didn't have the TPU parameters.)

* Address PR comments. Removed multi gpu flag + more

* fix lint

* fix more lints

* add todo for Synthetic dataset

* Update docs

* fix comment

* align comment

* add download script for librispeech dataset

* Apply changes.

* Apply changes

* apply comments

* update

* update file comment

* pylint

* update

* add progress bar, add more download flags

Added new model research/lm_commonsense

* Improve the BenchmarkFileLogger performance.

Open and preserve the file handler of the metric log file during
init, which reduce the overhead of open/close the file for each
log_metric call.

* Address review comment.

* add .take() to dataset pipeline

* delint

* address PR comments

* add tests for matmul embedding and schedule manager, as well as some minor cleanup

* delint

* address PR comments

Release evaluation code for OI Challenge 2018 and minor fixes.