GitLab

Adding validation every epoch allows us to view the progress during training instead of having to wait until the last eval. Mostly useful for manual runs.

Current batch size 160000 does not converge to the desired HR. So we decrease to 99k which is known to converge. Tested locally and got to 63.5 at epoch 7. Also decreasing number of epochs as I don't see any improvement after epoch 7-8.

249580533  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Internal change

249566870  by A. Unique TensorFlower<gardener@tensorflow.org>:

    Set up BERT benchmark test.

--

PiperOrigin-RevId: 249580533

* Add enable_get_next_as_optional flag.

* Set enable_get_next_as_optional to strategy.

* Add comments to explain the flag.

* Remove trailing whitespace.

* Remove trailing space.

249500988  by hongkuny<hongkuny@google.com>:

    Lints

--

PiperOrigin-RevId: 249500988

* 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

247226201  by ronnyvotel:

    Updating the visualization tools to accept unique_ids for color coding.

--
247067830  by Zhichao Lu:

    Add box_encodings_clip_range options for the convolutional box predictor (for TPU compatibility).

--
246888475  by Zhichao Lu:

    Remove unused _update_eval_steps function.

--
246163259  by lzc:

    Add a gather op that can handle ignore indices (which are "-1"s in this case).

--
246084944  by Zhichao Lu:

    Keras based implementation for SSD + MobilenetV2 + FPN.

--
245544227  by rathodv:

    Add batch_get_targets method to target assigner module to gather any groundtruth tensors based on the results of target assigner.

--
245540854  by rathodv:

    Update target assigner to return match tensor instead of a match object.

--
245434441  by Zhichao Lu:

    Add README for tpu_exporters package.

--
245381834  by lzc:

    Internal change.

--
245298983  by Zh...

* 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>:

  ...

249377254  by hongkuny<hongkuny@google.com>:

    Internal change

249373328  by hongkuny<hongkuny@google.com>:

    Clean up tf import

--
249333938  by hongkuny<hongkuny@google.com>:

    Fix tf1 import

--
249325089  by hongkuny<hongkuny@google.com>:

    BERT 2.0

--
249173564  by hongkuny<hongkuny@google.com>:

    Internal change

PiperOrigin-RevId: 249377254

* Initial feature aggregation code for Detect-to-Retrieve paper.

PiperOrigin-RevId: 246043144

* Add support for ASMK/ASMK*/R-ASMK/R-ASMK*.

PiperOrigin-RevId: 247337028

* Add DatumProto uint32 field, and limit datum_io to uint32 and float32/float64 types.

Also, introduce DatumPairProto, to be used for ASMK variants. Functions to read/write in this new format are added and tested.

PiperOrigin-RevId: 247515205

* Add batching option to feature aggregation extraction.

PiperOrigin-RevId: 247614627

* Script to perform local feature aggregation, with associated configs.

Also small edits to the aggregation extractor, for better handling of input features / avoiding OOM.

PiperOrigin-RevId: 248150750

* Tests to check that aggregation using regions with no local features works.

PiperOrigin-RevId: 248153275

* Include new library/proto for aggregation

* Merged commit includes the following changes:

PiperOrigin-RevId: 248176511

* Merged commit includes the following changes:
248194572  by Andre Araujo:

    Change tf.tensor_scatter_nd_add --> tf.compat.v1.tensor_scatter_add to make it compatible with TF 1.X.

--

PiperOrigin-RevId: 248194572

* Functions to parse ground-truth and compute metrics for revisited datasets.

Unit tests are added.

PiperOrigin-RevId: 248561575

* Small change to argparse bool option, which does not work as expected.

PiperOrigin-RevId: 248805505

* Class to compute similarity between aggregated descriptors.

PiperOrigin-RevId: 249102986

* Script to perform retrieval and compute metrics.

PiperOrigin-RevId: 249104011

* feature_aggregation_similarity library in DELF init

* D2R instructions / README update

* Small edit to README

* Internal change.

PiperOrigin-RevId: 249113531

* Instructions to reproduce D2R paper results, and small edits to config files.

PiperOrigin-RevId: 249159850

* Revert "Internal change."

This reverts commit 0529f2b8471a20c88f1fbb37367f142965d098ee.

Undoing incorrect markdown deletion.

* Small updates to markdown instructions

* Small updates to markdown instructions 2

* Initial feature aggregation code for Detect-to-Retrieve paper.

PiperOrigin-RevId: 246043144

* Add support for ASMK/ASMK*/R-ASMK/R-ASMK*.

PiperOrigin-RevId: 247337028

* Add DatumProto uint32 field, and limit datum_io to uint32 and float32/float64 types.

Also, introduce DatumPairProto, to be used for ASMK variants. Functions to read/write in this new format are added and tested.

PiperOrigin-RevId: 247515205

* Add batching option to feature aggregation extraction.

PiperOrigin-RevId: 247614627

* Script to perform local feature aggregation, with associated configs.

Also small edits to the aggregation extractor, for better handling of input features / avoiding OOM.

PiperOrigin-RevId: 248150750

* Tests to check that aggregation using regions with no local features works.

PiperOrigin-RevId: 248153275

* Include new library/proto for aggregation

* Merged commit includes the following changes:

PiperOrigin-RevId: 248176511

* Merged commit includes the following changes:
248194572  by Andre Araujo:

    Change tf.tensor_scatter_nd_add --> tf.compat.v1.tensor_scatter_add to make it compatible with TF 1.X.

--

PiperOrigin-RevId: 248194572

* Functions to parse ground-truth and compute metrics for revisited datasets.

Unit tests are added.

PiperOrigin-RevId: 248561575

* Small change to argparse bool option, which does not work as expected.

PiperOrigin-RevId: 248805505

* Class to compute similarity between aggregated descriptors.

PiperOrigin-RevId: 249102986

* Script to perform retrieval and compute metrics.

PiperOrigin-RevId: 249104011

* feature_aggregation_similarity library in DELF init

* D2R instructions / README update

* Small edit to README

* Internal change.

PiperOrigin-RevId: 249113531

* Instructions to reproduce D2R paper results, and small edits to config files.

PiperOrigin-RevId: 249159850

* Revert "Internal change."

This reverts commit 0529f2b8471a20c88f1fbb37367f142965d098ee.

Undoing incorrect markdown deletion.

* Initial feature aggregation code for Detect-to-Retrieve paper.

PiperOrigin-RevId: 246043144

* Add support for ASMK/ASMK*/R-ASMK/R-ASMK*.

PiperOrigin-RevId: 247337028

* Add DatumProto uint32 field, and limit datum_io to uint32 and float32/float64 types.

Also, introduce DatumPairProto, to be used for ASMK variants. Functions to read/write in this new format are added and tested.

PiperOrigin-RevId: 247515205

* Add batching option to feature aggregation extraction.

PiperOrigin-RevId: 247614627

* Script to perform local feature aggregation, with associated configs.

Also small edits to the aggregation extractor, for better handling of input features / avoiding OOM.

PiperOrigin-RevId: 248150750

* Tests to check that aggregation using regions with no local features works.

PiperOrigin-RevId: 248153275

* Include new library/proto for aggregation

* Merged commit includes the following changes:

PiperOrigin-RevId: 248176511

* Merged commit includes the following changes:
248194572  by Andre Araujo:

    Change tf.tensor_scatter_nd_add --> tf.compat.v1.tensor_scatter_add to make it compatible with TF 1.X.

--

PiperOrigin-RevId: 248194572

* Functions to parse ground-truth and compute metrics for revisited datasets.

Unit tests are added.

PiperOrigin-RevId: 248561575

* Small change to argparse bool option, which does not work as expected.

PiperOrigin-RevId: 248805505

* Class to compute similarity between aggregated descriptors.

PiperOrigin-RevId: 249102986

* Script to perform retrieval and compute metrics.

PiperOrigin-RevId: 249104011

* feature_aggregation_similarity library in DELF init

* D2R instructions / README update

* Small edit to README

* Internal change.

PiperOrigin-RevId: 249113531

* Instructions to reproduce D2R paper results, and small edits to config files.

PiperOrigin-RevId: 249159850

* Delete accuracy if exists in eval results.

* get global_step only if it exists in eval results

This will allow one to easily reproduce a benchmark by running with the flags.

* Set the --clone_model_in_keras_dist_strat to None.  Remove the separate no_cloning benchmarks and add a couple of cloning ones.  Fixes the learning rate schedule to cache its ops per graph.

* Added 'tfdata_exp' version of all benchmarks which set
FLAGS.tf_data_experimental_slack = True. Renamed
`data_prefetch_with_slack` to `data_delay_prefetch` (haoyu's change)
to make the names more distinct.

* Add flag to resnet input pipeline and surface through
keras_imagenet_main.py

* Initial feature aggregation code for Detect-to-Retrieve paper.

PiperOrigin-RevId: 246043144

* Add support for ASMK/ASMK*/R-ASMK/R-ASMK*.

PiperOrigin-RevId: 247337028

* Add DatumProto uint32 field, and limit datum_io to uint32 and float32/float64 types.

Also, introduce DatumPairProto, to be used for ASMK variants. Functions to read/write in this new format are added and tested.

PiperOrigin-RevId: 247515205

* Add batching option to feature aggregation extraction.

PiperOrigin-RevId: 247614627

* Script to perform local feature aggregation, with associated configs.

Also small edits to the aggregation extractor, for better handling of input features / avoiding OOM.

PiperOrigin-RevId: 248150750

* Tests to check that aggregation using regions with no local features works.

PiperOrigin-RevId: 248153275

* Include new library/proto for aggregation

* Merged commit includes the following changes:

PiperOrigin-RevId: 248176511

* Merged commit includes the following changes:
248194572  by Andre Araujo:

    Change tf.tensor_scatter_nd_add --> tf.compat.v1.tensor_scatter_add to make it compatible with TF 1.X.

--

PiperOrigin-RevId: 248194572

Create ssd_mobilenet_v2_fpnlite_quantized_shared_box_predictor_256x256_depthmultiplier_75_coco14_sync.config (#6775)

Add config for FPNLite quantized model

- Test passes locally python3 and test is already
    skipped for python2.

* 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.

* Fix trivial model to work properly with fp16

* Add comment on manual casting

Previously we had one dense layer in trivial model. The weight was [224*224*3, num_classes].
Using two dense layers, the weights are [224*224*3, 1] and [1, num_classes].

* Do not report metrics in performance benchmarks

* Rename flag

* Modified tweaked tests to use tensor learning rate

* 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.

* Add learning rate tensor. This makes training slower

* Improve LearningRateSchedule with better efficiency

* Fix lint error

* Replace constant definition with existing one

246873701  by menglong:

    Missing __init__.py under meta_architectures/

--
246857392  by menglong:

    Standardize proto namespace: lstm_object_detection.protos

--
246625127  by menglong:

    Internal changes.

--
246596481  by menglong:

    Add License

--
246580605  by menglong:

    Internal changes

--
246344626  by menglong:

    Open source interleaved mobilenet v2 model.

--
244893883  by menglong:

    Introduce multi_input_decoder for interleaved model.

--
244461016  by menglong:

    Add pre-bottleneck operation to lstm cells to support interleaved model.

--
244052176  by menglong:

    Update README

--
244020495  by menglong:

    Add test to rnn_decoder.

--
243704250  by menglong:

    Duplicate assignment.

--
243091836  by menglong:

    Move LSTMSSD meta arch into separate folder

--
242900337  by menglong:

    Modified mobilenet definition for LSTM-SSD

--
242...