GitLab

* Add new test ID and test env info to the benchmark run.

* Fix test.

* Fix lint

* Address review comment.

* Update benchmark logger to update the run status.

This is important for streaming upload to bigquery so that the
dashboard can ignore the 'running' benchmark at the moment since
its not finished yet.

* Move the run status into a separate table.

Also update the run status in the benchmark uploader and
BigqueryBenchmarkLogger.

* Insert instead of update for the benchmark status for file logger.

* Address review comments.

Update the logger to have benchmark context, which will update the
run status accordingly.

* Fix broken tests.

* Move the benchmark logger context to main function.

* Fix tests.

* Update the rest of the models to use the context in main.

* Delint.

* Fix hooks_test

* Add more comments

* Fix lints

* Exemplify csv handing in serving for boosted_trees model by using custom built signature_def.

* some minor touches.

* Reverted back to using the file instead of module in the example.

* Make boosted_trees Garden-official

* Fix nits

* Using BenchmarkLogger

* Using BenchmarkLogger

* Fixing tests

* Linting fixes.

* Adding comments

* Moving mock logger

* Moving mock logger

* Glinting

* Responding to CR

* Reverting assertEmpty

* Add unit test, official flags, and benchmark logs

* Fix checking errors

* Reorder imports to fix lints

* Address comments and correct model layers

* Add dataset checking

* Remove one hot labels, Add drop_remainder to batch, Use parallel interleve in imagenet dataset.

* minor lint fix

* Don't try to read the files twice...

* Add explanation for cycle_length

* Add recommendation model

* Fix pylints check error

* Rename file

* Address comments, update input pipeline, and add distribution strategy

* Fix import error

* Address more comments

* Fix lints

* Add boosted_trees to the official models

* Comments addressed from review, and a test added; using absl.flags instead of argparser.

* Used help_wrap. Also added instructions for inference.

* Fix resnet missing layers.

The official v1 model contains BN and Relu between input layer and
pooling.

* Remove the BN and Relu for V2.

After some discussion with team and refer to existing
implementation, those two layer seems to be only useful in V1.
In V2, the first unit of the block will have a projection, that
apply the BN and Relu for the shortcut. Adding a comment to make
this clear.

* Expand the comment section.

* Remove the pre-trained checkpoint since its broken right now.

Will restore it once we have new checkpoint generated.

* Update the wide_deep code for latest benchmark config.

* Also update the transformer benchmark code.

* Move the benchmark_uploader to new location.

* Update benchmark logger to streaming upload.

* Fix lint and unit test error.

* delint.

* Update the benchmark uploader test.

Skip the import of benchmark_uploader when bigquery is not installed.

* Merge the 2 classes of benchmark uploader into 1.

* Address review comments.

* delint.

* Execute bigquery upload in a separate thread.

* Change to use python six.moves for importing.

* Address review comments and delint.

* Address review comment.

Adding comment for potential performance impact for model on CPU.

* Fix random failure on py3.

* Fix the order of flag saver to avoid the randomness.

The test is broken when the benchmark_logger_type is set first, and
validated when the benchmark_log_dir is not set yet.

The final BN and ReLU layer is only need for v2 model since it was
doing preactivation in each block.

* Add benchmark logging for wide_deep.

* Fix lint error.

Update mnist_eager.py to use GPU if available and not specified otherwise. At the moment it uses the CPU even if a GPU is available, since "if flags_obj.no_gpu or tf.test.is_gpu_available():" in "def run_mnist_eager(flags_obj):" evaluates to True if a GPU is available.

* forbid resnet v1 fp16

* address PR comments

* rename --version flag and fix tests to correctly specify version rather than verbosity

* rename version to resnet_version throughout

* fix bugs

* delint

* missed layer_test

* fix indent

* Add dataset info and hyper parameter logging for benchmark.

* Address review comments.

* Address the view comment for data schema name.

* Fix test cases.

* Lint fix.

* Revert 823da318. This restores distribution strategies for resnet.

This commit is not a direct revert due to significant merge conflict
resolution.

* fix flags test

* npc is no longer used in resnet

* fix treatment of main() and absl flags

* add mnist_eager, and delint

* add mnist_eager docstring

* squash of modular absl usage commits

* delint

* address PR comments

* change hooks to comma separated list, as absl behavior for space separated lists is not as expected

* catch cpuinfo ImportError

* add psutil import catch

* fix typo

* Add new import paths to mnist_tpu.py

This adds new import paths to mnist_tpu.py so individuals can check out the git repository and run `python mnist_tpu.py` without finagling with python import search paths.

* Fix lint errors

* Add disable= to pylint instructions

* Remove parent path

* Update comments

Refer to commit "use reduce_mean instead of Average Pooling"(d5663b3c), "second_pool_size" and "second_pool_stride" no longer needed in resnet_model

1. trainning hooks and train/eval function does not need to be
declared several times.
2. change to use tf.logging.info instead of print
3. Show the current index and total count of training cycle, which
give user some hint about where they are in the whole process.

Previously the timestamp we pushed to bigquery was PDT, but the
timezone spec was set to 'Z' (UTC). This is provide incorrect
value to bigquery, and might affect analysis down the road.