Commit Graph

138 Commits

Author SHA1 Message Date
Jiuqiang Tang
3bc8276678 Remove "All Rights Reserved." in copyright headers.
PiperOrigin-RevId: 526982992
2023-04-25 09:06:29 -07:00
MediaPipe Team
56df724c36 Add customizable face stylizer module in MediaPipe model maker
PiperOrigin-RevId: 526883862
2023-04-25 00:47:45 -07:00
MediaPipe Team
33c8c68bba Add a default_applicable_licenses to model_maker/python/vision/core.
PiperOrigin-RevId: 526716940
2023-04-24 12:08:34 -07:00
MediaPipe Team
abded49e5b Internal change
PiperOrigin-RevId: 526300079
2023-04-22 10:52:14 -07:00
MediaPipe Team
a6c1bb6324 Internal change
PiperOrigin-RevId: 526235882
2023-04-22 00:28:52 -07:00
MediaPipe Team
3e0ed2ced0 Internal Changes
PiperOrigin-RevId: 525180095
2023-04-18 10:14:08 -07:00
MediaPipe Team
43fd744296 Internal Changes
PiperOrigin-RevId: 524997017
2023-04-17 17:43:40 -07:00
MediaPipe Team
0067a1b5c2 Internal changes
PiperOrigin-RevId: 522248624
2023-04-05 22:37:11 -07:00
MediaPipe Team
5a1a9269e6 Internal Changes
PiperOrigin-RevId: 522247775
2023-04-05 22:31:13 -07:00
MediaPipe Team
7ae4d0175a CL will fix the typos in the tasks files
PiperOrigin-RevId: 522240681
2023-04-05 21:42:19 -07:00
MediaPipe Team
6605f551e7 Object Detector add batch_size and train_data to get_steps_per_epoch.
PiperOrigin-RevId: 522149938
2023-04-05 14:11:29 -07:00
MediaPipe Team
5615c1e459 Delete duplicate public APIs in object detector
PiperOrigin-RevId: 522098326
2023-04-05 10:58:08 -07:00
MediaPipe Team
d9f940f8b2 Model Maker object detector change learning_rate_boundaries to learning_rate_epoch_boundaries.
PiperOrigin-RevId: 521024056
2023-03-31 15:19:26 -07:00
MediaPipe Team
88d68341de Migrate face stylizer model files to GCS to use downloadable models in model maker.
PiperOrigin-RevId: 520848629
2023-03-31 00:11:51 -07:00
MediaPipe Team
1eeb89e95f Add the model configuration and training hyperparameters for BlazeFaceStylizer.
PiperOrigin-RevId: 520767282
2023-03-30 16:07:11 -07:00
MediaPipe Team
0c8f691a36 Internal changes
PiperOrigin-RevId: 520411557
2023-03-29 12:35:20 -07:00
MediaPipe Team
9ee02481b1 Fix object detector pascal voc dataloader issue
PiperOrigin-RevId: 520160549
2023-03-28 16:06:44 -07:00
MediaPipe Team
97c271644a Open Source Object Detector
PiperOrigin-RevId: 519201221
2023-03-24 11:53:18 -07:00
MediaPipe Team
27a5a6d433 Internal Changes
PiperOrigin-RevId: 518976200
2023-03-23 15:20:50 -07:00
MediaPipe Team
cb2ee87705 Internal change
PiperOrigin-RevId: 518591192
2023-03-22 09:34:23 -07:00
MediaPipe Team
560945ad39 Internal Changes
PiperOrigin-RevId: 517219631
2023-03-16 13:58:02 -07:00
MediaPipe Team
3b66ac0623 Import the saved Keras models of BlazeFaceStylizer components into MediaPipe model maker.
PiperOrigin-RevId: 517044265
2023-03-16 00:44:06 -07:00
MediaPipe Team
6774794d02 Add the dataset module for face stylizer in model maker.
PiperOrigin-RevId: 516628350
2023-03-14 14:11:20 -07:00
MediaPipe Team
c94de4032d Fix preprocess Callable typing
PiperOrigin-RevId: 515818356
2023-03-10 21:50:46 -08:00
MediaPipe Team
2d8f937913 Improve docstring of image classifier model spec.
PiperOrigin-RevId: 515466722
2023-03-09 16:06:40 -08:00
Yuqi Li
9e7950a69a Make each model file in the model asset bundle to be aligned relative to the start of the file (alignment = 4).
PiperOrigin-RevId: 511624410
2023-02-22 16:15:24 -08:00
MediaPipe Team
40b0dc960a Internal Changes
PiperOrigin-RevId: 511604893
2023-02-22 14:55:03 -08:00
MediaPipe Team
bdd1c24990 Refactor common methods into vision/core/image_utils.py and vision/core/test_utils.py
PiperOrigin-RevId: 509968910
2023-02-15 16:54:38 -08:00
MediaPipe Team
99fc975f49 Add download model on demand to text classifier
PiperOrigin-RevId: 508441452
2023-02-09 12:15:17 -08:00
MediaPipe Team
28f728bed5 Change gesture_recognizer to download model files on-demand from GCS.
PiperOrigin-RevId: 508424508
2023-02-09 11:13:06 -08:00
MediaPipe Team
d283e6a05a Support downloading model files on-demand from GCS in model_maker
PiperOrigin-RevId: 506174708
2023-01-31 18:43:35 -08:00
MediaPipe Team
afb0182935 Internal model maker change.
PiperOrigin-RevId: 504472342
2023-01-24 23:16:21 -08:00
MediaPipe Team
4e135ccdb9 Internal Model Maker change.
PiperOrigin-RevId: 504315641
2023-01-24 10:38:25 -08:00
MediaPipe Team
9cbb76939d Adds smaller MobileBERT model.
PiperOrigin-RevId: 501451414
2023-01-11 20:35:39 -08:00
MediaPipe Team
d40fa6b16d Internal Model Maker change.
PiperOrigin-RevId: 500758488
2023-01-09 11:04:45 -08:00
MediaPipe Team
e3131d7d78 Internal change
PiperOrigin-RevId: 499521620
2023-01-04 10:33:04 -08:00
MediaPipe Team
add5600d0d Internal change
PiperOrigin-RevId: 499351795
2023-01-03 17:20:46 -08:00
MediaPipe Team
7e36a5e2ae Set filecmp.cmp(shallow=False) in model_maker unit tests.
PiperOrigin-RevId: 498218578
2022-12-28 11:25:28 -08:00
MediaPipe Team
5a71b551e5 Remove duplicate and non-public api for model_maker
PiperOrigin-RevId: 497251246
2022-12-22 15:31:12 -08:00
MediaPipe Team
36f054dfbe Internal model maker change
PiperOrigin-RevId: 497196512
2022-12-22 10:43:15 -08:00
MediaPipe Team
5c0f548f5f Switches to tf.keras.optimizers.experimental.AdamW instead of the legacy AdamW.
PiperOrigin-RevId: 496821354
2022-12-20 20:53:13 -08:00
MediaPipe Team
a7b52d2c52 Internal changes
PiperOrigin-RevId: 496754449
2022-12-20 14:37:18 -08:00
MediaPipe Team
3e6cd5d2bf Add support for customizing gesture recognizer layers
PiperOrigin-RevId: 496456160
2022-12-19 11:57:09 -08:00
MediaPipe Team
b4e1969e43 Add pip package builder for model_maker
PiperOrigin-RevId: 493989013
2022-12-08 14:03:15 -08:00
MediaPipe Team
a641ea12e1 Update gesture recognizer to new mediapipe tasks pipeline
PiperOrigin-RevId: 493950564
2022-12-08 11:32:42 -08:00
MediaPipe Team
02aa162c9e Rename gesture_recognizer test_data to testdata to be consistent with rest of model_maker
PiperOrigin-RevId: 492246728
2022-12-01 11:13:19 -08:00
Scott Zhu
a367753eda Internal change
PiperOrigin-RevId: 489824381
2022-11-20 10:41:20 -08:00
MediaPipe Team
bdf4078e89 Internal change
PiperOrigin-RevId: 489752009
2022-11-19 21:14:24 -08:00
MediaPipe Team
977ee4272e Add public visibility to the model maker public API.
PiperOrigin-RevId: 489701768
2022-11-19 10:53:21 -08:00
MediaPipe Team
524ac3ca61 Internal change for Model Maker
PiperOrigin-RevId: 489540387
2022-11-18 12:48:30 -08:00
MediaPipe Team
1fb0902aa0 Update gesture_recognizer test
PiperOrigin-RevId: 489301508
2022-11-17 14:04:30 -08:00
MediaPipe Team
74474d859e Update image_classifier demo with new ImageClassifierOption changes
PiperOrigin-RevId: 489031381
2022-11-16 14:00:39 -08:00
MediaPipe Team
cdd44e77b7 Internal change
PiperOrigin-RevId: 488969539
2022-11-16 10:07:40 -08:00
MediaPipe Team
2f77bf44e3 Use train_data to evaluate accuracy of unit test for gesture_recognizer due to limited dataset size.
PiperOrigin-RevId: 488808942
2022-11-15 18:10:28 -08:00
MediaPipe Team
e65f21e2d8 Update the docstring to make it consistent with the model option update.
PiperOrigin-RevId: 488761331
2022-11-15 14:36:48 -08:00
MediaPipe Team
1689112b23 Improve model_util_test code.
PiperOrigin-RevId: 488752497
2022-11-15 14:04:39 -08:00
MediaPipe Team
f14645cb06 Model maker gesture recognizer test changes
PiperOrigin-RevId: 488702055
2022-11-15 10:50:32 -08:00
MediaPipe Team
aafbf73c0a Renames model options in TextClassifier.
PiperOrigin-RevId: 488063797
2022-11-12 10:02:19 -08:00
MediaPipe Team
c2ac040a6c Adds a public import API for TextClassifier.
PiperOrigin-RevId: 487949023
2022-11-11 16:52:16 -08:00
MediaPipe Team
3e05871f98 Open source Model Maker text tasks.
PiperOrigin-RevId: 487706929
2022-11-10 19:52:51 -08:00
MediaPipe Team
ec327cedcb Internal Changes
PiperOrigin-RevId: 487673720
2022-11-10 16:48:13 -08:00
MediaPipe Team
a5bcb97d88 Adds an AverageWordVecModel.
PiperOrigin-RevId: 487104909
2022-11-08 18:29:21 -08:00
MediaPipe Team
c31aaa94a6 Adds a BertClassifier.
PiperOrigin-RevId: 487086744
2022-11-08 16:49:46 -08:00
MediaPipe Team
0917e8cb8e Support continual training image classifier from saved checkpoint files.
PiperOrigin-RevId: 487057612
2022-11-08 14:50:04 -08:00
MediaPipe Team
b3d19fa1af Use model bundle writer when exporting models in gesture recognizer
PiperOrigin-RevId: 487042776
2022-11-08 13:52:08 -08:00
MediaPipe Team
4a6562d423 internal
PiperOrigin-RevId: 486853812
2022-11-07 22:39:23 -08:00
MediaPipe Team
24d03451c7 Internal change - migration
PiperOrigin-RevId: 486853689
2022-11-07 22:36:22 -08:00
MediaPipe Team
c3bb4bb5da Refactor image classifier Create API by defining an ImageClassifierOption which wraps the model_spec, model_options, and hparams. Also migrate the definition of HParams by extending it from BaseHParams.
PiperOrigin-RevId: 486796657
2022-11-07 16:39:35 -08:00
MediaPipe Team
6b54cae34c Internal change - cleanup
PiperOrigin-RevId: 486721059
2022-11-07 11:42:50 -08:00
MediaPipe Team
b2cc2cc60c Internal changes
PiperOrigin-RevId: 486014305
2022-11-03 17:23:20 -07:00
MediaPipe Team
d29c3d7512 Add metadata writer to image_classifier in model_maker
PiperOrigin-RevId: 485926985
2022-11-03 11:30:38 -07:00
MediaPipe Team
0bf5110a7e Add a model_maker internal library directory.
PiperOrigin-RevId: 485665201
2022-11-02 12:31:58 -07:00
MediaPipe Team
459214e6a3 Rename index_by_label to label_names.
PiperOrigin-RevId: 484956259
2022-10-30 22:26:17 -07:00
MediaPipe Team
a28c9d2c26 Set steps_per_epoch to None when calling model.fit() method for image classifier.
PiperOrigin-RevId: 483764377
2022-10-25 13:44:03 -07:00
MediaPipe Team
ec2a34d2a4 Replace pytype_struct_contrib_test by py_strict_test. Also remove unnecessary BUILD attributes.
PiperOrigin-RevId: 483237371
2022-10-23 16:45:02 -07:00
MediaPipe Team
e453c3b206 Migrates image_preprocessing from core to vision.
PiperOrigin-RevId: 482856336
2022-10-21 12:16:54 -07:00
MediaPipe Team
55ba23ce9a Cleans up unused function from image_classifier's Dataset.
PiperOrigin-RevId: 482703775
2022-10-20 23:59:20 -07:00
MediaPipe Team
348c4e6652 Add a test to test saving and loading the gesture recognizer model.
Also update the loss function init method to resolve a bug when loading model.

PiperOrigin-RevId: 482702020
2022-10-20 23:45:26 -07:00
MediaPipe Team
086fc442fd Internal change
PiperOrigin-RevId: 482697999
2022-10-20 23:15:17 -07:00
MediaPipe Team
e71638cf67 Add an export_tflite API to gesture recognizer model maker library.
PiperOrigin-RevId: 482527017
2022-10-20 10:21:34 -07:00
MediaPipe Team
3d588bae8b Internal change
PiperOrigin-RevId: 482259130
2022-10-19 11:34:32 -07:00
MediaPipe Team
f2821d840d Improve image classifier model maker documentation and replace the legacy code.
PiperOrigin-RevId: 481979922
2022-10-18 11:47:37 -07:00
MediaPipe Team
51879ae81a Remove unnecessary architectures from image classifier ModelSpec
PiperOrigin-RevId: 481974529
2022-10-18 11:29:11 -07:00
MediaPipe Team
0428550d75 Open source model_maker/python/core/tasks and model_maker/python/vision/image_classifier
PiperOrigin-RevId: 481182271
2022-10-14 10:47:34 -07:00
MediaPipe Team
1b611c66bb Improve quantization support in model_maker/image_classifier
PiperOrigin-RevId: 480455944
2022-10-11 14:40:02 -07:00
MediaPipe Team
f4fd1063a7 Add helper methods to load saved model from external files in model maker.
PiperOrigin-RevId: 480444918
2022-10-11 13:58:05 -07:00
MediaPipe Team
6fa455f40e Dataclasses for text classifier
PiperOrigin-RevId: 480178697
2022-10-10 14:33:31 -07:00
MediaPipe Team
fb3b0e788e Open sourcing model_maker/core/util and model_maker/core/data
PiperOrigin-RevId: 478835650
2022-10-04 11:10:53 -07:00