GitLab

* Merged commit includes the following changes:
244869387  by Sergio Guadarrama:

    This CL adds script/code to generate Visual WakeWords Dataset annotation files and TF records starting from COCO dataset.

--
244866660  by Sergio Guadarrama:

    Add VisualWakeWords Dataset to Slim dataset_factory to train MobileNets on it.

--
244836000  by Sergio Guadarrama:

    n/a

--
244104396  by Sergio Guadarrama:

    Add an option whether to enable / disable image cropping in inception_preprocessing.

--
242040128  by Sergio Guadarrama:

    Internal change

241793677  by Sergio Guadarrama:

    Internal change

241073081  by Sergio Guadarrama:

    Internal change

240131189  by Sergio Guadarrama:

    Internal change

PiperOrigin-RevId: 244869387

* Merged commit includes the following changes:
245431876  by Sergio Guadarrama:

    Internal cleanup

--

PiperOrigin-RevId: 245431876

* Merged commit includes the following changes:
245454983  by Sergio Guadarrama:

    Internal Cleanup

--

PiperOrigin-RevId: 245454983

233991726  by Sergio Guadarrama:

    Internal change

231925959  by Sergio Guadarrama:

    Internal change

231253502  by Sergio Guadarrama:

    Internal change

229973546  by Sergio Guadarrama:

    Internal change

229870842  by Sergio Guadarrama:

    Internal change

PiperOrigin-RevId: 233991726

228203246  by Sergio Guadarrama:

    Add a write text graphdef option.

--
226110161  by Sergio Guadarrama:

    Add license to i3d/s3dg and tests.

--
226074013  by Sergio Guadarrama:

    Network definitions for I3D and S3D-G.

--
224394404  by Sergio Guadarrama:

    Add video model option for exported inference graphs.

--
224220779  by Sergio Guadarrama:

    Internal change

223589268  by Sergio Guadarrama:

    Internal change

PiperOrigin-RevId: 228203246

223150784  by Sergio Guadarrama:

    Allow using batch norm scale parameters for Inception models.

--
221391590  by Sergio Guadarrama:

    Add support for group normalization in the object detection API. Just adding MobileNet-v1 SSD currently. This may serve as a road map for other models that wish to support group normalization as an option.

--
221342582  by Sergio Guadarrama:

    Internal change

220817084  by Sergio Guadarrama:

    Internal change

216005108  by Sergio Guadarrama:

    Introduce hparam `use_bounded_activation` for NASNet. The hparam decides whether to use
    1. bounded activation
    2. clip_by_value for the add operands and bounded activation after add operator.
    3. bounded activation before 'none' and 'pooling' branch
    The restriction on the tensor value range makes it compatible with quantized inference.

--

PiperOrigin-RevId: 223150784

* Merged commit includes the following changes:
213899768  by Sergio Guadarrama:

    Fixes #3819.

--
213493831  by Sergio Guadarrama:

    Internal change

212057654  by Sergio Guadarrama:

    Internal change

210747685  by Sergio Guadarrama:

    For FPN, when use_depthwise is set to true, use slightly modified mobilenet v1 config.

--
210128931  by Sergio Guadarrama:

    Allow user-defined current_step in NASNet.

--
209092664  by Sergio Guadarrama:

    Add quantized fine-tuning / training / eval and export to slim image classifier binaries.

--
207651347  by Sergio Guadarrama:

    Update mobilenet v1 docs to include revised tflite models.

--
207165245  by Sergio Guadarrama:

    Internal change

207095064  by Sergio Guadarrama:

    Internal change

PiperOrigin-RevId: 213899768

* Update model_lib.py to fix eval_spec name issue.

* PiperOrigin-RevId: 201234832

* PiperOrigin-RevId: 202507333

* PiperOrigin-RevId: 204320344

* Add PNASNet-5 mobile network model and cell structure.

PiperOrigin-RevId: 204735410

* Add option to customize individual projection layer activation.

PiperOrigin-RevId: 204776951

2. Flag that allows to prevent imagenet.py  from downloading label_to_names from github and/or dumping into training directory (which might be read-only)
3. Adds some comments about how decay steps are computed, since it computed differently when there are clones vs sync replicas.
4. Updates mobilenet.md to describe the training process using train_image_classifer
5. Add citation for TF-Slim model library.

PiperOrigin-RevId: 191955231

PiperOrigin-RevId: 193254125

PiperOrigin-RevId: 193371562

PiperOrigin-RevId: 194085628

PiperOrigin-RevId: 194857067

PiperOrigin-RevId: 196125653

PiperOrigin-RevId: 196589070

PiperOrigin-RevId: 199522873

PiperOrigin-RevId: 200351305

* Make codes more pythonic.

* Restore the indents

Restore the indents.

* Adding option for one_box_for_all_classes to the box_predictor

PiperOrigin-RevId: 192813444

* Extend to accept different ratios of conv channels.

PiperOrigin-RevId: 192837477

* Remove inaccurate caveat from proto file.

PiperOrigin-RevId: 192850747

* Add option to set dropout for classification net in weight shared box predictor.

PiperOrigin-RevId: 192922089

* fix flakiness in testSSDRandomCropWithMultiClassScores due to randomness.

PiperOrigin-RevId: 193067658

* Post-process now works again in train mode.

PiperOrigin-RevId: 193087707

* Adding support for reading in logits as groundtruth labels and applying an optional temperature (scaling) before softmax in support of distillation.

PiperOrigin-RevId: 193119411

* Add a util function to visualize value histogram as a tf.summary.image.

PiperOrigin-RevId: 193137342

* Do not add batch norm parameters to final conv2d ops that predict boxes encodings and class scores in weight shared conv box predictor.

This allows us to set proper bias and force initial predictions to be background when using focal loss.

PiperOrigin-RevId: 193204364

* Make sure the final layers are also resized proportional to conv_depth_ratio.

PiperOrigin-RevId: 193228972

* Remove deprecated batch_norm_trainable field from ssd mobilenet v2 config

PiperOrigin-RevId: 193244778

* Updating coco evaluation metrics to allow for a batch of image info, rather than a single image.

PiperOrigin-RevId: 193382651

* Update protobuf requirements to 3+ in installation docs.

PiperOrigin-RevId: 193409179

* Add support for training keypoints.

PiperOrigin-RevId: 193576336

* Fix data augmentation functions.

PiperOrigin-RevId: 193737238

* Read the default batch size from config file.

PiperOrigin-RevId: 193959861

* Fixing a bug in the coco evaluator.

PiperOrigin-RevId: 193974479

* num_gt_boxes_per_image and num_det_boxes_per_image value incorrect.
Should be not the expand dim.

PiperOrigin-RevId: 194122420

* Add option to evaluate any checkpoint (without requiring write access to that directory and overwriting any existing logs there).

PiperOrigin-RevId: 194292198

* PiperOrigin-RevId: 190346687

* - Expose slim arg_scope function to compute keys to enable tessting.
- Add is_training=None option to mobinenet arg_scopes. This allows the users to set is_training from an outer scope.

PiperOrigin-RevId: 190997959

* Add an option to not set slim arg_scope for batch_norm is_training parameter. This enables users to set the is_training parameter from an outer scope.

PiperOrigin-RevId: 191611934

* PiperOrigin-RevId: 191955231

* PiperOrigin-RevId: 193254125

* PiperOrigin-RevId: 193371562

* PiperOrigin-RevId: 194085628

* fix issue 'could not satisfy explicit device'

* remove the line unrelated to fix issue

* PiperOrigin-RevId: 189857068

* PiperOrigin-RevId: 190089200

* Merge pull request #3702 from cclauss/from-six.moves-import-xrange-yet-again

from six.moves import xrange (en masse) YET AGAIN

PiperOrigin-RevId: 190255581

* I Fixes bunch of model tests that were using python2 functions.

II Updates mobilenet code:
1) Mobilenet usage example
2) Links to all checkpoints and updated README
3) Performance graphs

PiperOrigin-RevId: 190300379

* PiperOrigin-RevId: 190306214

* Updates notebook to reflect canonical repository location and fixes few
variable names.

* Force cast of num_classes to integer

PiperOrigin-RevId: 188335318

* Updating config util to allow overwriting of cosine decay learning rates.

PiperOrigin-RevId: 188338852

* Make box_list_ops.py and box_list_ops_test.py work with C API enabled.

The C API has improved shape inference over the original Python
code. This causes some previously-working conds to fail. Switching to smart_cond fixes this.

Another effect of the improved shape inference is that one of the
failures tested gets caught earlier, so I modified the test to reflect
this.

PiperOrigin-RevId: 188409792

* Fix parallel event file writing issue.

Without this change, the event files might get corrupted when multiple evaluations are run in parallel.

PiperOrigin-RevId: 188502560

* Deprecating the boolean flag of from_detection_checkpoint.

Replace with a string field fine_tune_checkpoint_type to train_config to provide extensibility. The fine_tune_checkpoint_type can currently take value of `detection`, `classification`, or others when the restore_map is overwritten.

PiperOrigin-RevId: 188518685

* Automated g4 rollback of changelist 188502560

PiperOrigin-RevId: 188519969

* Introducing eval metrics specs for Coco Mask metrics. This allows metrics to be computed in tensorflow using the tf.learn Estimator.

PiperOrigin-RevId: 188528485

* Minor fix to make object_detection/metrics/coco_evaluation.py python3 compatible.

PiperOrigin-RevId: 188550683

* Updating eval_util to handle eval_metric_ops from multiple `DetectionEvaluator`s.

PiperOrigin-RevId: 188560474

* Allow tensor input for new_height and new_width for resize_image.

PiperOrigin-RevId: 188561908

* Fix typo in fine_tune_checkpoint_type name in trainer.

PiperOrigin-RevId: 188799033

* Adding mobilenet feature extractor to object detection.

PiperOrigin-RevId: 188916897

* Allow label maps to optionally contain an explicit background class with id zero.

PiperOrigin-RevId: 188951089

* Fix boundary conditions in random_pad_to_aspect_ratio to ensure that min_scale is always less than max_scale.

PiperOrigin-RevId: 189026868

* Fallback on from_detection_checkpoint option if fine_tune_checkpoint_type isn't set.

PiperOrigin-RevId: 189052833

* Add proper names for learning rate schedules so we don't see cryptic names on tensorboard.

PiperOrigin-RevId: 189069837

* Enforcing that all datasets are batched (and then unbatched in the model) with batch_size >= 1.

PiperOrigin-RevId: 189117178

* Adding regularization to total loss returned from DetectionModel.loss().

PiperOrigin-RevId: 189189123

* Standardize the names of loss scalars (for SSD, Faster R-CNN and R-FCN) in both training and eval so they can be compared on tensorboard.

Log localization and classification losses in evaluation.

PiperOrigin-RevId: 189189940

* Remove negative test from box list ops test.

PiperOrigin-RevId: 189229327

* Add an option to warmup learning rate in manual stepping schedule.

PiperOrigin-RevId: 189361039

* Replace tf.contrib.slim.tfexample_decoder.LookupTensor with object_detection.data_decoders.tf_example_decoder.LookupTensor.

PiperOrigin-RevId: 189388556

* Force regularization summary variables under specific family names.

PiperOrigin-RevId: 189393190

* Automated g4 rollback of changelist 188619139

PiperOrigin-RevId: 189396001

* Remove step 0 schedule since we do a hard check for it after cl/189361039

PiperOrigin-RevId: 189396697

* PiperOrigin-RevId: 189040463

* PiperOrigin-RevId: 189059229

* PiperOrigin-RevId: 189214402

* Force regularization summary variables under specific family names.

PiperOrigin-RevId: 189393190

* Automated g4 rollback of changelist 188619139

PiperOrigin-RevId: 189396001

* Make slim python3 compatible.

* Monir fixes.

* Add TargetAssignment summaries in a separate family.

PiperOrigin-RevId: 189407487

* 1. Setting `family` keyword arg prepends the summary names twice with the same name. Directly adding family suffix to the name gets rid of this problem.
2. Make sure the eval losses have the same name.

PiperOrigin-RevId: 189434618

* Minor fixes to make object detection tf 1.4 compatible.

PiperOrigin-RevId: 189437519

* Call the base of mobilenet_v1 feature extractor under the right arg scope and set batchnorm is_training based on the value passed in the constructor.

PiperOrigin-RevId: 189460890

* Automated g4 rollback of changelist 188409792

PiperOrigin-RevId: 189463882

* Update object detection syncing.

PiperOrigin-RevId: 189601955

* Add an option to warmup learning rate, hold it constant for a certain number of steps and cosine decay it.

PiperOrigin-RevId: 189606169

* Let the proposal feature extractor function in faster_rcnn meta architectures return the activations (end_points).

PiperOrigin-RevId: 189619301

* Fixed bug which caused masks to be mostly zeros (caused by detection_boxes being in absolute coordinates if scale_to_absolute=True.

PiperOrigin-RevId: 189641294

* Open sourcing Mobilenetv2 + SSDLite.

PiperOrigin-RevId: 189654520

* Remove unused files.

Filename was wrong

* Internal change.

PiperOrigin-RevId: 187042423

* Internal change.

PiperOrigin-RevId: 187072380

* Opensource float and eight-bit fixed-point mobilenet_v1 training and eval scripts.

PiperOrigin-RevId: 187106140

* Initial check-in for Mobilenet V2

PiperOrigin-RevId: 187213595

* Allow configuring batch normalization decay and epsilon in MobileNet v1

PiperOrigin-RevId: 187425294

* Allow overriding NASNet model HParams.

This is a change to the API that will allow users to pass in their own configs
to the building functions, which should make these APIs much more customizable
for end-user cases.

This change removes the use_aux_head argument from the model construction
functions, which is no longer necessary given that the use_aux_head option is
configurable in the model config. For example, for the mobile ImageNet model,
the auxiliary head can be disabled using:

config = nasnet.mobile_imagenet_config()
config.set_hparam('use_aux_head', 0)
logits, endpoints = nasnet.build_nasnet_mobile(
    inputs, num_classes, config=config)
PiperOrigin-RevId: 188617685

* Automated g4 rollback of changelist 188617685

PiperOrigin-RevId: 188619139

* Removes spurious comment

Some python files under models/research, it is wrong to use "Rather then" instead of "Rather than" in  comments.

* BUG: Fix inception imagenet download script

In `download_imagenet.sh`, the path to the `$SYNSETS_FILE` resource
(imagenet_2012_validation_synset_labels.txt) was not updated to
reflect change in directory after a `cd` earlier in the script.

Fixes #682.

* BUG: Fix slim imagenet download script

In `download_imagenet.sh`, the path to the `$SYNSETS_FILE` resource
(imagenet_2012_validation_synset_labels.txt) was not updated to
reflect change in directory after a `cd` earlier in the script.

* Merged commit includes the following changes:
186565198  by Sergio Guadarrama:

    Applied random_hsv_in_yiq in inception_preprocessing.

--
186501039  by Sergio Guadarrama:

    Applied random_hsv_in_yiq in inception_preprocessing.

--
186013907  by Sergio Guadarrama:

    Internal change

185715309  by Sergio Guadarrama:

    Obviates the need for prepadding on mobilenet v1 and v2 for fully convolutional models.

--
184266252  by Sergio Guadarrama:

    Give build_nasnet_*() functions an optional flag use_aux_head,
    and add an internal-only arg scope to NasNetA*Cell._apply_drop_path().

--
183865228  by Sergio Guadarrama:

    Internal change

179580924  by Sergio Guadarrama:

    Internal change

177320302  by Sergio Guadarrama:

    Internal change

177130184  by Sergio Guadarrama:

    Make slim nets tests faster by using smaller examples of oversized inputs.

--
176965289  by Sergio Guadarrama:

    Internal change

176585260  by Sergio Guadarrama:

    Internal change

176534973  by Sergio Guadarrama:

    Internal change

175526881  by Sergio Guadarrama:

    Internal change

174967704  by Sergio Guadarrama:

    Treat num_classes=0 same as None in a few slim nets overlooked by the recent
    change.

--
174443227  by Sergio Guadarrama:

    Internal change

174281864  by Sergio Guadarrama:

    Internal change

174249903  by Sergio Guadarrama:

    Fix nasnet image classification and object detection by moving the option to turn ON or OFF batch norm training into it's own arg_scope used only by detection

--
173954505  by Sergio Guadarrama:

    Merge pull request #2651 from sguada/tmp1

    Fixes imports

    Closes #2636

    ORIGINAL_AUTHOR=Jon Shlens <shlens@users.noreply.github.com>
    COPYBARA_INTEGRATE_REVIEW=https://github.com/tensorflow/models/pull/2636 from tensorflow:sguada-patch-1 19ff570f52df5ab655c00fb439129b201c5f2dce

--
173928094  by Sergio Guadarrama:

    Remove pending imports

--

PiperOrigin-RevId: 186565198

* Remove internal links.

1) Fix word dir
2) Fix download url

Replaces #2183  Python 2.7 no longer needs the trailing L and Python 3 treats it as a Syntax Error.

tf.contrib.framework.get_or_create_global_step -> tf.train.get_or_create_global_step

Fix nasnet image classification and object detection by moving the
option to turn ON or OFF batch norm training into it's own arg_scope
used only by detection

PiperOrigin-RevId: 173917628