GitLab

PiperOrigin-RevId: 335952206

PiperOrigin-RevId: 335944731

PiperOrigin-RevId: 335759679

* Merged commit includes the following changes:
326369548  by Andre Araujo:

    Fix import issues.

--
326159826  by Andre Araujo:

    Changed the implementation of the cosine weights from Keras layer to tf.Variable to manually control for L2 normalization.

--
326139082  by Andre Araujo:

    Support local feature matching using ratio test.

    To allow for easily choosing which matching type to use, we rename a flag/argument and modify all related files to avoid breakages.

    Also include a small change when computing nearest neighbors for geometric matching, to parallelize computation, which saves a little bit of time during execution (argument "n_jobs=-1").

--
326119848  by Andre Araujo:

    Option to measure DELG latency taking binarization into account.

--
324316608  by Andre Araujo:

    DELG global features training.

--
323693131  by Andre Araujo:

    PY3 conversion for delf public lib.

--
321046157  by Andre Araujo:

    Purely Google refactor

--

PiperOrigin-RevId: 326369548

* Added export of delg_model module.

* README update to explain training DELG global features head

* Added guidelines for DELF hyperparameter values

* Fixed typo

* Added mention about remaining training flags.

* Merged commit includes the following changes:
334723489  by Andre Araujo:

    Backpropagate global and attention layers together.

--
334228310  by Andre Araujo:

    Enable scaling of local feature locations to the resized resolution.

--

PiperOrigin-RevId: 334723489
Co-authored-by: Andre Araujo <andrearaujo@google.com>

PiperOrigin-RevId: 335692122

to support both model and encoder as key in ckpt when exporting to tfhub

PiperOrigin-RevId: 335668823

Update the default arguments for orbit.Controller train_and_evaluate() and evaluate_continuously() to match the default for evaluate().

PiperOrigin-RevId: 335659533

Add it to export. TODO: move the encoder and decoder layer out to layers/ or decoders/.

PiperOrigin-RevId: 335560860

PiperOrigin-RevId: 335548133

PiperOrigin-RevId: 335521712

Imported from GitHub PR https://github.com/tensorflow/tensorflow/pull/42625

Miss `training` call argument for dropout layer. /cc @tanzhenyu for visibility.
Copybara import of the project:

--
ba2198fc735a2deca08c68a3c6266d02e01dfe3b by Tzu-Wei Sung <windqaq@gmail.com>:

Add training call argument for MultiHeadAttention

Remove trailing space

PiperOrigin-RevId: 335507127

PiperOrigin-RevId: 335446217

PiperOrigin-RevId: 335226408

PiperOrigin-RevId: 335128787

PiperOrigin-RevId: 335115606

PiperOrigin-RevId: 335106919

PiperOrigin-RevId: 335079988

PiperOrigin-RevId: 335069348

PiperOrigin-RevId: 335067840

PiperOrigin-RevId: 335066419

PiperOrigin-RevId: 335049970

Remove type annotation for params.
Make the trainer consume both model and task.

PiperOrigin-RevId: 334964198

PiperOrigin-RevId: 334950562

PiperOrigin-RevId: 334942484

PiperOrigin-RevId: 334910093

PiperOrigin-RevId: 334860469

PSv2: Replace existing `tf.distribute.experimental.ParameterServerStrategy` usage with `tf.compat.v1.distribute.experimental.ParameterServerStrategy` to prepare for the upcoming TF2 ParameterServerStrategy API release.

The practically only difference for API endpoint switch is the monitoring from V2 to V1, for those who were using `tf.distribute.experimental.ParameterServerStrategy`. It's not supported in V2 and should be tracked as V1 anyway.

PiperOrigin-RevId: 334847114

PiperOrigin-RevId: 334829363

Changes output_tfrecord_path_prefix to output_tfrecord_path.

PiperOrigin-RevId: 334817429

PiperOrigin-RevId: 334647537

PiperOrigin-RevId: 334642892

PiperOrigin-RevId: 334528962

PiperOrigin-RevId: 334517779

PiperOrigin-RevId: 334233893

PiperOrigin-RevId: 334085862

PiperOrigin-RevId: 334062074

PiperOrigin-RevId: 333979065

PiperOrigin-RevId: 333949668