GitLab

PiperOrigin-RevId: 411136341

All models which support loss scaling support dynamic loss scaling, so the argument has no purpose. It used to be that some models scaled the loss manually instead of using a LossScaleOptimizer, and so did not support dynamic loss scaling.

PiperOrigin-RevId: 367719521

PiperOrigin-RevId: 331359058

PiperOrigin-RevId: 326286926

Remove is_v2_0

PiperOrigin-RevId: 312336907

Profiler has been integrated with TensorBoard callback and released in TF 2.2

PiperOrigin-RevId: 311475711

PiperOrigin-RevId: 308451074

PiperOrigin-RevId: 307421021

PiperOrigin-RevId: 305897677

The TF pip packages are always built with CUDA support, so tf.test.is_built_with_cuda() would return True even if the user had no GPU.

PiperOrigin-RevId: 302928378

PiperOrigin-RevId: 301185629

PiperOrigin-RevId: 300461275

PiperOrigin-RevId: 300247220

PiperOrigin-RevId: 299160422

PiperOrigin-RevId: 297298712

PiperOrigin-RevId: 296988935

PiperOrigin-RevId: 292007895

PiperOrigin-RevId: 291810091

PiperOrigin-RevId: 291432304

utils/misc/keras_utils

PiperOrigin-RevId: 283885102

PiperOrigin-RevId: 282293668

PiperOrigin-RevId: 266458583

Moving Keras ResNet models to `official/vision/image_classification` and benchmarks to `official/benchmark`.

PiperOrigin-RevId: 264268533

* force_v2_in_keras_compile FLAG default to None and added seperate temp path.

* switch to force testing 1v path not force v2 path.

* Rename function force_v1_path.

261339941  by haoyuzhang<haoyuzhang@google.com>:

    Own library functions in Keras ResNet models, and remove dependencies on v1 Estimator version of ResNet models.

    Most dependencies that the Keras version has are related to data input pipelines. Created dedicated files (cifar_preprocessing.py, imagenet_preprocessing.py) to collect all logic handling Cifar and ImageNet data input function.

--
261339166  by haoyuzhang<haoyuzhang@google.com>:

    Internal change

261317601  by akuegel<akuegel@google.com>:

    Internal change

261218818  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

PiperOrigin-RevId: 261339941

* Add force_run_distributed tests.

* Added enable_eager

* r/force_run_distributed/force_v2_in_keras_compile

* Adding force_v2 tests and FLAGs.

* Rename method to avoid conflict.

* Add cpu force_v2 tests.

* fix lint, wrap line.

* change to force_v2_in_keras_compile

* Update method name.

* Lower mlperf target to 0.736.

* Config threadpool, cuDNN persistent BN, and grappler layout optimizer properly for ResNet56

* Add tweaked tests for Resnet56

* Avoid triggering the last partial batch overhead by explicitly dropping remainder

* Improve performance for Cifar ResNet benchmarks

* Revert batch size changes to benchmarks

* XLA FP32 and first test

* More XLA benchmarks FP32.

* Add eager to NCF and refactor resnet.

* fix v2_0 calls and more flag refactor.

* Remove extra flag args.

* 90 epoch default

* add return

* remove xla not used by estimator.

* Remove duplicate run_eagerly.

* fix flag defaults.

* Remove fp16_implementation flag option.

* Remove stop early on mlperf test.

* remove unneeded args.

* load flags from keras mains.

* Do not set learning phase when skipping eval

* Do not set learning phase in no dist strat case

* Added device placement, tweaked benchmarks

* Added tweaked benchmarks for Cifar

* Fix device scope

* Fix lint

* Add explicit GPU placement flag

* Also run accuracy test with explicit GPU placement

* Added doc string

* set default steps to 300K.

* Log flags to perfzero.

* Add XLA support to transformer

- Moved config logic to keras_utils
- Added enable_xla flag to _performance flags
- Did not refactor enable_xla flag from keras resnet due to
  reliance on calling FLAGs in estimator keras and that is
  a needed refactor for another time.

* fix g3 lint complaint.

* Refactor set config into keras_utils.

* Move flags out of main.

* pipe through enable_xla

* Update official/transformer/v2/misc.py
Co-Authored-By: Reed <reedwm@google.com>

* add ds support for ncf

* remove comments for in_top_k

* avoid expanding the input layers

* resolve comments and fix lint

* Added some comments in code and fix lint

* fix lint

* add some documentation

* add tensorflow imports

* Add profiler callback for Keras models

* Update build stats to identify time callback by type

* Add warning message when both TensorBoard and profiler callbacks are used