GitLab

* Fix various lint errors

* Fix logging format

* Make max_length and static_batch configurable.

* Fix line length.

* Fix incorrect parameters in building eval input.

* Improve comments for readability.

* Add static batch benchmarks to estimator 

So we can distinguish how much static vs dynamic batch matter.

* change max_length for static_batch tests

* Add flag for max length

this is not going to help with current tf.data semantics. so removing it.

* Fixes that make transformer run.

* Remove debug print statements.

* Changed the permissions to 644.

* Fix the rest of the permissions.

* enable static batch in all benchmarks

* Restrict dist strat hack to training mode

For now we will do predict/eval without dist strat, so remove that hack in non training cases.

* Use `inputs` instead of `x` as arg name for call

Keras has different behavior based on whether the inputs are called `inputs` or not. Using `inputs` gives expected behaviors.

* Avoid extra map fn on input in dist strat case

* Update how we handle custom metrics

This new approach works with and without dist strat. The previous one didn't work with dist strat. We need to fix that but this is reasonable in meantime (b/133724664).

* Update benchmarks

* typo in metrics code

* Revert metrics change

Didn't actually work in distributed case..

This shuffling should help in getting shuffling each epoch.

* Moved common keras code to utils.

* Initial 1 gpu benchmark

- Aligned flags with resnet example
- removed code/features that are not super useful
- eval as part of train if bleu source/ref provided
- add exp_per_second hook

* Rename benchmark classes, pass batch-size and log_steps.

* fix docstring

* Predict done with checkpoints inline

- perfzero baseclass

* steps not epochs with smoother training loop.

* do not initialize history outside loop.

* 5000 between eval not 500

* estimator to keras.

* remove epochs var.

* use range not xrange.

* 200K steps for 1 gpu

* fix global step

* Merged commit includes the following changes:
249776315  by tianlin<tianlin@google.com>:

    Internal change

249763206  by tianlin<tianlin@google.com>:

    For TF 2.0 (related to Beam Search), expand cond dims in tf.where(cond, x, y) to make all parameters broadcastable.

--
249392724  by hongkuny<hongkuny@google.com>:

    Internal change

PiperOrigin-RevId: 249776315

* Merged commit includes the following changes:
249823043  by tianlin<tianlin@google.com>:

    Bring back v2 test for predict and eval.

--

PiperOrigin-RevId: 249823043

* Add big tests.

* fix super

* Add fp16, increase 8xGPU batch-sizes

* Adding the rest of the fp16 tests.

* Big accuracy test batch_perf_gpu

* fix docstrings

* add _run_and_report

* Edited docstrings

* Merged commit includes the following changes:
249218656  by tianlin<tianlin@google.com>:

    Deal with imports, fix a typo and make unit tests fast.

--
249198645  by tianlin<tianlin@google.com>:

    Trivial: Remove one empty line before "import tensorflow"

--
249195490  by tianlin<tianlin@google.com>:

    Initialize Transformer TF V2 Model with Keras subclassing implementation. (Compatible with TF V1)

--
249195008  by tianlin<tianlin@google.com>:

    Internal change

249173564  by hongkuny<hongkuny@google.com>:

    Internal change

249079258  by hongkuny<hongkuny@google.com>:

    Internal change

247691534  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

247533725  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

247509295  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

247311355  by wangtz<wangtz@google.com>:

    Internal change

247303127  by wangtz<wangtz@google.com>:

    Minor changes

--
247277315  by wangtz<wangtz@google.com>:

    Internal change

247143663  by wangtz<wangtz@google.com>:

    Internal change

247129668  by wangtz<wangtz@google.com>:

    Internal change

--
247110547  by wangtz<wangtz@google.com>:

    Internal change

247016722  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

246677582  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

245821839  by shiningsun<shiningsun@google.com>:

    Internal change

245353681  by gjn<gjn@google.com>:

    Internal change

245340898  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

245291749  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

245155641  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

244962751  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

244395199  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

244098932  by shiningsun<shiningsun@google.com>:

    Internal change

244019160  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

242965373  by shiningsun<shiningsun@google.com>:

    Internal change

242930998  by shiningsun<shiningsun@google.com>:

    Internal change

242914617  by shiningsun<shiningsun@google.com>:

    Internal change

242510000  by shiningsun<shiningsun@google.com>:

    Internal change

242183356  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

242049350  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241749800  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241663771  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241416315  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241054800  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

241028555  by yuefengz<yuefengz@google.com>:

    Internal change

240683076  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

240404735  by yuefengz<yuefengz@google.com>:

    Internal change

239892001  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

239316550  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

238316437  by shiningsun<shiningsun@google.com>:

    Internal change

238251867  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

237876559  by taylorrobie<taylorrobie@google.com>:

    Internal change

236346619  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

236182665  by tayo<tayo@google.com>:

    Internal change

235035787  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

234652747  by wangtz<wangtz@google.com>:

    Internal change

234606543  by wangtz<wangtz@google.com>:

    Internal change

233837502  by shiningsun<shiningsun@google.com>:

    Internal change

232033015  by shiningsun<shiningsun@google.com>:

    Internal change

232028127  by shiningsun<shiningsun@google.com>:

    Internal change

232026986  by shiningsun<shiningsun@google.com>:

    Internal change

232024086  by shiningsun<shiningsun@google.com>:

    Internal change

229839087  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

229794563  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

229204024  by shiningsun<shiningsun@google.com>:

    Internal change

228956426  by shiningsun<shiningsun@google.com>:

    Internal change

228610870  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

228586727  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

228564809  by taylorrobie<taylorrobie@google.com>:

    Internal change

227080093  by shiningsun<shiningsun@google.com>:

    Internal change

227052580  by shiningsun<shiningsun@google.com>:

    Internal change

226217872  by gsundeep<gsundeep@google.com>:

    Internal change

225436264  by shiningsun<shiningsun@google.com>:

    Internal change

222283824  by taylorrobie<taylorrobie@google.com>:

    Internal change

219719958  by taylorrobie<taylorrobie@google.com>:

    Internal change

219330498  by taylorrobie<taylorrobie@google.com>:

    Internal change

219241224  by taylorrobie<taylorrobie@google.com>:

    Internal change

218774474  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

218610966  by taylorrobie<taylorrobie@google.com>:

    Internal change

218576353  by taylorrobie<taylorrobie@google.com>:

    Internal change

217815929  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

217776707  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

217749789  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

217029365  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

216803510  by taylorrobie<taylorrobie@google.com>:

    Internal change

216570979  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

215640939  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

214516790  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

212485310  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

212339556  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

210935187  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

210658133  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

207194426  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

206963562  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

206866123  by taylorrobie<taylorrobie@google.com>:

    Internal change

206631416  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

206180517  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

206170536  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

205252141  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

204191308  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

202519641  by scottzhu<scottzhu@google.com>:

    Internal change

201577585  by scottzhu<scottzhu@google.com>:

    Internal change

201299684  by kathywu<kathywu@google.com>:

    Internal change

199825697  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

199655516  by karmel<karmel@google.com>:

    Internal change

199654555  by karmel<karmel@google.com>:

    Internal change

199653009  by karmel<karmel@google.com>:

    Internal change

199209802  by karmel<karmel@google.com>:

    Internal change

198089630  by karmel<karmel@google.com>:

    Internal change

198064765  by karmel<karmel@google.com>:
    Automated rollback of changelist 197740116.

198060863  by karmel<karmel@google.com>:
    Automated rollback of changelist 197920496.

197920496  by kathywu<kathywu@google.com>:

    Internal change

197841416  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

197740116  by kathywu<kathywu@google.com>:

    Internal change

196608994  by karmel<karmel@google.com>:

    Internal change

196172336  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

195867348  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

195725348  by taylorrobie<taylorrobie@google.com>:

    Internal change

195283704  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

194960508  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

194662698  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

194103064  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

193735742  by saeta<saeta@google.com>:

    Support legacy clusters

--
193581866  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

193567921  by taylorrobie<taylorrobie@google.com>:

    Internal change

193072964  by taylorrobie<taylorrobie@google.com>:

    Internal change

192783651  by scottzhu<scottzhu@google.com>:
    Automated rollback of changelist 192714881.

192714881  by scottzhu<scottzhu@google.com>:
    Automated rollback of changelist 192710755.

192710755  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

192374551  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

192346754  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

192298443  by karmel<karmel@google.com>:

    Internal change

192220576  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191937900  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191909212  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191602380  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191514106  by scottzhu<scottzhu@google.com>:

    Internal change

191355033  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

191327699  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

190938103  by karmel<karmel@google.com>:

    Internal change

190804388  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

190780502  by karmel<karmel@google.com>:

    Internal change

190479716  by karmel<karmel@google.com>:

    Internal change

190289991  by karmel<karmel@google.com>:

    Internal change

189844661  by scottzhu<scottzhu@google.com>:
    Automated rollback of changelist 189816818.

189816818  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

189796731  by kathywu<kathywu@google.com>:

    Internal change

189639056  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

189628781  by karmel<karmel@google.com>:

    Internal change

189267175  by karmel<karmel@google.com>:

    Internal change

189222334  by karmel<karmel@google.com>:

    Internal change

189096159  by karmel<karmel@google.com>:

    Internal change

189085341  by karmel<karmel@google.com>:

    Internal change

188949700  by karmel<karmel@google.com>:

    Internal change

188780170  by karmel<karmel@google.com>:

    Internal change

188525399  by karmel<karmel@google.com>:

    Internal change

188189681  by karmel<karmel@google.com>:

    Internal change

188076840  by wun<wun@google.com>:

    Internal change

188074937  by wun<wun@google.com>:

    Internal change

187764140  by wun<wun@google.com>:

    Internal change

187663993  by wun<wun@google.com>:

    Internal change

187651035  by karmel<karmel@google.com>:

    Internal change

187415223  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

187047094  by saeta<saeta@google.com>:

    Integrate ClusterResolvers with TPUEstimator.

--
186843813  by karmel<karmel@google.com>:

    Internal change

186843458  by karmel<karmel@google.com>:

    Internal change

186348279  by karmel<karmel@google.com>:

    Internal change

186323602  by ashankar<ashankar@google.com>:

    Internal change

185051452  by wun<wun@google.com>:

    Internal change

185039308  by wun<wun@google.com>:

    Internal change

184934877  by wun<wun@google.com>:

    Internal change

184876413  by wun<wun@google.com>:

    Internal change

184868771  by wun<wun@google.com>:

    Internal change

184751017  by wun<wun@google.com>:

    Internal change

184743769  by wun<wun@google.com>:

    Internal change

183894402  by frankchn<frankchn@google.com>:

    Internal change

183863555  by frankchn<frankchn@google.com>:

    Internal change

183249343  by karmel<karmel@google.com>:

    Internal change

183099733  by wun<wun@google.com>:

    Internal change

182981253  by wun<wun@google.com>:

    Internal change

182977703  by wun<wun@google.com>:

    Internal change

182414330  by ashankar<ashankar@google.com>:

    Internal change

182069924  by wun<wun@google.com>:

    Internal change

181544879  by ashankar<ashankar@google.com>:

    Internal change

180618064  by ashankar<ashankar@google.com>:

    Internal change

180591779  by ashankar<ashankar@google.com>:

    Internal change

179724765  by kathywu<kathywu@google.com>:

    Internal change

179713599  by kathywu<kathywu@google.com>:

    Internal change

179110125  by wun<wun@google.com>:

    Internal change

178793339  by kathywu<kathywu@google.com>:

    Internal change

176581733  by wun<wun@google.com>:

    Internal change

175220569  by wun<wun@google.com>:

    Internal change

175183195  by wun<wun@google.com>:

    Internal change

175171381  by wun<wun@google.com>:

    Internal change

175036833  by wun<wun@google.com>:

    Internal change

175031858  by wun<wun@google.com>:

    Internal change

174891336  by wun<wun@google.com>:

    Internal change

174881248  by kathywu<kathywu@google.com>:

    Internal change

174766157  by wun<wun@google.com>:

    Internal change

173723629  by wun<wun@google.com>:

    Internal change

173579695  by wun<wun@google.com>:

    Internal change

173443901  by wun<wun@google.com>:

    Internal change

173340656  by wun<wun@google.com>:

    Internal change

172954249  by wun<wun@google.com>:

    Internal change

172926654  by wun<wun@google.com>:

    Internal change

172922540  by wun<wun@google.com>:

    Internal change

172921684  by kathywu<kathywu@google.com>:

    Internal change

170943723  by wun<wun@google.com>:

    Internal change

170927898  by wun<wun@google.com>:

    Internal change

170430173  by wun<wun@google.com>:

    Internal change

170428312  by wun<wun@google.com>:

    Internal change

PiperOrigin-RevId: 249218656

* Revert "Merged commit includes the following changes:"

This reverts commit e30d805d3770d7473166dcca1ecb6050dc5de47a.

* Merged commit includes the following changes:

PiperOrigin-RevId: 249232280

* Merged commit includes the following changes:
249376444  by tianlin<tianlin@google.com>:

    Internal change

249373328  by hongkuny<hongkuny@google.com>:

    Clean up tf import

--
249340512  by hongkuny<hongkuny@google.com>:

    Internal change

249325089  by hongkuny<hongkuny@google.com>:

    BERT 2.0

--
249260134  by tianlin<tianlin@google.com>:

    Internal change

PiperOrigin-RevId: 249376444

* Fixed absl_app in official/mnist/mnist_tpu.py

* Make metrics in order (switch from dict to list of tuples).

* Revert ResNet Model to an earlier version.

* Fix lint format.

* Disable two tests in v2.

* Add FP16 and benchmarks.

* add missing run and report.

* Add loss_scale as option not included with dtype.

* move loss_scale validation under dtype conditional.

* add loss_scale to flags tested.

* Add first benchmark and return stats.

* Remove print statements update training steps.

* Revert print T: in print statement.

* Remove print(stats)

* add 2 gpu accuracy test for base.

* Fixed total_batch_size when using gpu + gFile deprecations.

* 8 GPU test name fix

* Add 4 and 8 GPU tests.

* typo fixes.

* Clean up test names and methods.

* bleu uncased.  docstring format fix.

Replace per_device with per_replica and PerDevice with PerReplica, because the PerDevice concept was renamed and doesn't exist anymore. (#6693)

* Replace per_device with per_replica and PerDevice with PerReplica, because the PerReplica concept was renamed and doesn't exist anymore.

* Update README.md

* Update README.md

* Update README.md

* Add a flag to specify distribution strategies.

* Fix a small error.

* Address comments.

* Address comments.

* Fix typos.

It currently fails with
TypeError: not all arguments converted during string formatting

For tf2 this will only be available in `compat.v1`.

Explicitly pass values kwarg to tf.name_scope as it is currently being treated as the default_name kwarg instead. This causes an exception to be thrown in eager mode.

* set strip_default_attrs=True for SavedModel exports

* specify dtype in resnet export

* another dtype fix

* fix another dtype issue, and set --image_bytes_as_serving_input to default to False

Bypassing Export model step, if training on TPU's. As this need inference to be supported on TPU's. Remove this check once inference is supported. (#5209)

Bypassing Export model step, if training on TPU's. As this need inference to be supported on TPU's. Remove this check once inference is supported.

* Deterministic dataset order fix

In order for the order of the files to be deterministic, in `tf.data.Dataset.list_files(..., shuffle)`, shuffle needs to be True, otherwise different iterator inits will yield different file orders

* removed unnecessary shuffle of filenames

* Removed the `_FILE_SHUFFLE_BUFFER` definition

* Fix Transformer TPU crash in Python 2.X.

- Tensorflow raises an error when tf_inspect.getfullargspec is called on
a functools.partial in Python 2.X. This issue would be hit during the
eval stage of the Transformer TPU model. This change replaces the call
to functools.partial with a lambda to work around the issue.

* Remove unused import from transformer_main.

* Fix lint error.