GitLab

* Added TF-Lite-compatible feature extractor and model exporter for YAMNet.

- Added a TF-Lite compatible feature extractor. With the latest TF-Lite,
  that involves a DFT-multiplication replacement for tf.abs(tf.signal.stft())
  and not a lot else. Note that TF-Lite now allows variable-length inputs.
- Added a YAMNet exporter that produces TF2 SavedModels, TF-Lite models,
  and TF-JS models.
- Cleanups: switched hyperparameters to a dataclass, got rid of
  some lingering cruft in yamnet_test.

* Responded to DAn's comments in https://github.com/tensorflow/models/pull/9098

- Switched some hparams to float
- Made class map asset available on the exported model, and tested that
  it can be loaded from the various exports.

PiperOrigin-RevId: 327114257

PiperOrigin-RevId: 327094300

PiperOrigin-RevId: 327072144

PiperOrigin-RevId: 326947485

PiperOrigin-RevId: 326947209

PiperOrigin-RevId: 326935344

PiperOrigin-RevId: 326825957

PiperOrigin-RevId: 326739356

PiperOrigin-RevId: 326738853

Testing: Trained Bert2Bert model with MLIR bridge enabled and then NHNet based on Bert2Bert checkpoint, also with MLIR bridge enabled. Exported this to a saved model, ran inference, and confirmed model performance, accuracy, and prediction parity with the model trained without the MLIR bridge enabled.
PiperOrigin-RevId: 326734770

PiperOrigin-RevId: 326731205

* Merged commit includes the following changes:
326723075  by Andre Araujo:

    Move image resize utility into utils.py.

--

PiperOrigin-RevId: 326723075

* Adding back matched_images_demo.png

PiperOrigin-RevId: 326722720

Update to TF 2 for deep_speech

i.e., the checkpoint will be saved using
tf.train.Checkpoint(FLAGS.checkpoint_model_name=model)

PiperOrigin-RevId: 326672697

PiperOrigin-RevId: 326665731

PiperOrigin-RevId: 326604435

PiperOrigin-RevId: 326586473

* 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.
Co-authored-by: Andre Araujo <andrearaujo@google.com>

PiperOrigin-RevId: 326550493

PiperOrigin-RevId: 326541553

PiperOrigin-RevId: 326541072

PiperOrigin-RevId: 326538981

PiperOrigin-RevId: 326534159

PiperOrigin-RevId: 326528933

PiperOrigin-RevId: 326500498

PiperOrigin-RevId: 326496940

PiperOrigin-RevId: 326495091

PiperOrigin-RevId: 326483694

PiperOrigin-RevId: 326396920