experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
104,463 Commits 49 Branches 133 Tags 692 MiB
mm-llvm-bazel
Commit Graph

881 Commits

Author SHA1 Message Date
A. Unique TensorFlower
985ad0276a PY2 removal cleanup 
PiperOrigin-RevId: 352907145
Change-Id: I82de30d92dc9c2b53215d6d5732c67afe339c23d
 2021-01-20 17:11:44 -08:00
A. Unique TensorFlower
034633f23b PY2 removal cleanup 
PiperOrigin-RevId: 352106691
Change-Id: I382d53c64f0d29da430b8cb6d2395a2cb281509e
 2021-01-15 16:48:57 -08:00
Mark Daoust
08af2ba27c Add a public readme explaining the state of this examples directory. 
PiperOrigin-RevId: 339454688
Change-Id: If60bc8c5783d09769d828386932d84cf4bf065cd
 2020-10-28 07:29:47 -07:00
Mark Daoust
c8a9751c55 Remove saved_model test from examples. 
PiperOrigin-RevId: 336914830
Change-Id: Ib15d58225c837d9550901eddc623961be028cac7
 2020-10-13 11:23:37 -07:00
A. Unique TensorFlower
f5236aa296 BUILD file cleanup 
PiperOrigin-RevId: 334697600
Change-Id: I4f22e8c9a3dcd073f037847c8fde99e5ec1f3a35
 2020-09-30 16:03:47 -07:00
Mark Daoust
0947a774d0 Move deprecated example. 
from examples/android to tools/android/test.

All changes are just find/replace the path.

PiperOrigin-RevId: 334371911
Change-Id: If90dc1ddf8d11fdf71a6c9f9c77072cb08736c5a
 2020-09-29 07:19:05 -07:00
Mark Daoust
2fa2b797d0 Delete ios example. 
PiperOrigin-RevId: 334195144
Change-Id: I031102fb1e8a3837ab3956e78d6f397659f15714
 2020-09-28 11:12:22 -07:00
Mark Daoust
42e122e6a0 Remove examples/tutorials. Examples have moved to their own repo. 
PiperOrigin-RevId: 333794392
Change-Id: I5496b0d1673e8317660dbc2c5140c47220d4c468
 2020-09-25 13:26:26 -07:00
Mark Daoust
c31acd156c Remove example, and old redirect. 
Prefer https://tensorflow.org/tutorials and https://github.com/tensorflow/docs

PiperOrigin-RevId: 333549270
Change-Id: I37d6805bd6eef1d5bdb29c25d0549cbba7aacf52
 2020-09-24 11:04:05 -07:00
Mark Daoust
efb9be47d2 remove empty directory. 
PiperOrigin-RevId: 333535662
Change-Id: Icfe9f1195ca3e9a067fdb01d42defa76c345b437
 2020-09-24 10:10:26 -07:00
TensorFlower Gardener
2856a636b1 Merge pull request from raven38:raven38-patch 
PiperOrigin-RevId: 328273735
Change-Id: I57b78630da1dd4482114fdba7a2a6e2493d83b08
 2020-08-24 22:56:53 -07:00
Kai Katsumata
829248cb74 Fix typo in mnist/input_data.py 2020-08-23 00:18:49 +09:00
A. Unique TensorFlower
e8598ce045 Break up tensorflow/core/kernels/BUILD (part 3 of N): 
Move image processing kernels to subdirectory tensorflow/core/kernels/image with its own BUILD file.

PiperOrigin-RevId: 326056277
Change-Id: Ibe8813eeb36432c6220bdd14de40898fdef195f2
 2020-08-11 11:04:52 -07:00
Ran Chen
e1e2645945 Put distribution and cluster parameter modifiers in their own combination 
It's easier to add more strategy/cluster modifiers.

PiperOrigin-RevId: 324859132
Change-Id: Ided1a3d1106d71e86335ca3c59d7698550fa2155
 2020-08-04 12:02:29 -07:00
Hannes Achleitner
2fee5bb856 fix manifest lint error 2020-07-23 06:25:51 +02:00
Hannes Achleitner
8deeb07947 targetSdkVersion 29 2020-07-23 06:25:51 +02:00
Hannes Achleitner
0efc20fc94 Update libraries 2020-07-23 06:12:28 +02:00
Hannes Achleitner
4771a3f138 ignore .cxx directory 2020-07-23 06:11:59 +02:00
Hannes Achleitner
2889981f87 Android Studio 4.0 2020-07-23 05:44:22 +02:00
Hannes Achleitner
4bc2773024 Update Gradle 2020-07-23 05:43:59 +02:00
Hannes Achleitner
ce62fb78de Give Android sample a meaningfull name 2020-07-23 05:40:29 +02:00
Gaurav Jain
016bb3fc99 Remove unnecessary eval() calls 
The assertAll* statements already evaluate the arguments.

PiperOrigin-RevId: 319130109
Change-Id: I0034bdfa87a974613561a39e0d4a4223292245c7
 2020-06-30 17:18:32 -07:00
Frank Chen
4f2d979e1c Routine dead code clean-up. 
PiperOrigin-RevId: 315385914
Change-Id: I5c3be3a2b221d65197a375dbee99731ddc130438
 2020-06-08 17:20:27 -07:00
Revan Sopher
9cd1555bd2 Disable saved_mode_test on Cloud TPU. 
PiperOrigin-RevId: 313835836
Change-Id: Idbc66938bad01cf41dd042c0e05551cd79db8717
 2020-05-29 14:08:14 -07:00
Ran Chen
d29f1d6fe6 Declare tests that uses distribute/combinations as Python3 
I'll soon introduce PY3 only dependency to distribute/combinations to better
simulate multi worker.

PiperOrigin-RevId: 313669514
Change-Id: Ia986971a7fee469002614a9e272c475e9f93a250
 2020-05-28 15:32:21 -07:00
A. Unique TensorFlower
5d3c548620 Resolve trivial aliases for portable TensorFlow targets. 
PiperOrigin-RevId: 311548335
Change-Id: I837aa5a62500682783607841f0c993c2b6c238ed
 2020-05-14 09:39:36 -07:00
A. Unique TensorFlower
87856fd39d Revert skipping part of saved_model_test.test_mnist_cnn. 
PiperOrigin-RevId: 308238967
Change-Id: I536f0a0b86b351d5bc64ff2b6b0af2bb239b96d3
 2020-04-24 05:46:47 -07:00
Rick Chao
3e2ea0fcbe Skip part of saved_model_test.test_mnist_cnn due to failure on guitar. 
PiperOrigin-RevId: 308150234
Change-Id: I4378e567274a7860bd555652959ffa29c063537d
 2020-04-23 16:21:56 -07:00
Pete Warden
314b503bd2 Add SavedModel support to speech commands training 
PiperOrigin-RevId: 307691321
Change-Id: Ib52d1bf5537835ce56db11b074a0f3df9c3f9206
 2020-04-21 15:24:55 -07:00
Amit Patankar
056c2058c3 Internal fixes for code coverage testing. 
PiperOrigin-RevId: 306261564
Change-Id: I8a2a1e867036f4678812025a504dfa684693c3a4
 2020-04-13 10:51:10 -07:00
Jeremy Lau
0a3f4e1533 Internal fixes for code coverage testing. 
PiperOrigin-RevId: 305981529
Change-Id: I8d96fbc9a4ba66c652c509b4c247dd199753f1f6
 2020-04-10 18:31:29 -07:00
Ran Chen
64f4a59d5e Add multi worker mirrored strategy combinations to enable easier testing again 
The previous was rolled back because it breaks windows builds.

The change adds has_chief and num_workers to NamedDistribution. If there're more than one workers (chief+workers), the distribute flavor of combinations library will run the test in multiple processes to simulate a multi worker setup. Users are required to call combinations.main() instead of test.main().

Note that it's the same as running the test concurrently in multiple processes. You're expected to program your test in the same as if you're writing a multi client program. There's no way to get the return value from all processes. If you need that, use multi_process_runner directly.

This is very slow at this moment. Do not use it for a large number of tests.

PiperOrigin-RevId: 305136390
Change-Id: I814323cccaee5b8b7eed3b4dfef2971ab9d09cb4
 2020-04-06 15:59:19 -07:00
Brian Zhao
bd74bba60d Automated g4 rollback of changelist 304516887. 
PiperOrigin-RevId: 304711859
Change-Id: I98b1b95612fe0aad9689e6fcd460334b105ab6bd
 2020-04-03 16:09:34 -07:00
Ran Chen
2883cb5fb1 Add multi worker mirrored strategy combinations to enable easier testing 
It's required to use it with @run_in_each_worker.

When a test method is annotated with @run_in_each_worker, we call
multi_process_runner.run() under the hood. The test method is thus executed on
*each* worker, and is not executed in the original process. The original process
will assert on the success of the test on all workers.

This is very slow at this moment. Do not use it for a large number of tests.

PiperOrigin-RevId: 304516887
Change-Id: I6006cb53354c00f5f2cbedf94ff843a498bce5b4
 2020-04-02 18:15:41 -07:00
Angus-Luo
9fc04fc531 fix pylint errors (Line too long) 2020-03-27 11:51:14 +08:00
Angus-Luo
652a4b64be tf.SparseTensor to tf.sparse.SparseTensor 2020-03-25 10:49:19 +08:00
TensorFlower Gardener
fa3451c646 Merge pull request from foldl:fix-float-wav-data 
PiperOrigin-RevId: 302427524
Change-Id: I3d9201bce1f454e855a23553a6a1b1223cb398e0
 2020-03-23 07:22:24 -07:00
TensorFlower Gardener
f46ac90866 Merge pull request from Angus-Luo:master 
PiperOrigin-RevId: 301172077
Change-Id: Id2a2d3dd49f198fb463eef7838d38d62c49adb6b
 2020-03-16 09:33:07 -07:00
angusluo
d21998f3eb
change tf.image.decode_image to tf.io.decode_image 
tf.image.decode_image to tf.io.decode_image for API migration and consistency.
Please see https://www.tensorflow.org/api_docs/python/tf/io/decode_image
 2020-03-13 12:02:42 +08:00
A. Unique TensorFlower
76de54cc83 Remove uses of inline_constants on android_binary from BUILD files. 
PiperOrigin-RevId: 298487539
Change-Id: I93912357c3fbdec8a0ddfff88b2f7a4efa0e43ed
 2020-03-02 17:32:25 -08:00
Ken Franko
dac3721584 Remove TODO since workaround makes sense. 
PiperOrigin-RevId: 297535246
Change-Id: I4bb3006bd42b0d637d6517049b7e6f16c4f4e63b
 2020-02-27 00:57:49 -08:00
Meghna Natraj
3fd71fca2c Update speech_commands example: Code cleanup by resolving warnings and errors. 
PiperOrigin-RevId: 296559221
Change-Id: I07442f9296dd02f2a6ef46969a3efc475c2235e5
 2020-02-21 20:07:46 -08:00
TensorFlower Gardener
23b4405ace Merge pull request from foldl:make-speech-command-v2-compat 
PiperOrigin-RevId: 296056103
Change-Id: Ibee2e8731ff7e8ebf19e5fa4e53f62d95634e4b8
 2020-02-19 14:23:08 -08:00
Meghna Natraj
d9c9c92c7c Fix bug in dropout. 
PiperOrigin-RevId: 295776879
Change-Id: Ic25abd0fe0e442f37a32c7f68307e43728658b71
 2020-02-18 11:57:30 -08:00
Judd
e390647355
Update accuracy_utils.py 
fix pylint warning/errors.
 2020-02-14 16:06:02 +08:00
Judd
8e3fc97982
Update test_streaming_accuracy.py 
fix v2 compatibility.
 2020-02-14 14:36:43 +08:00
Judd
19ecdb017a
Update accuracy_utils.py 
fix v2 compatibility.
 2020-02-14 14:33:34 +08:00
Meghna Natraj
9b8390678f Automated rollback of commit 3d6df0985d 
PiperOrigin-RevId: 294940942
Change-Id: I31c21b47627016e8d037b30da552ef40c6ec508d
 2020-02-13 10:10:10 -08:00
Meghna Natraj
615b6c01dc Fix warnings and minor refactoring. 
PiperOrigin-RevId: 294698198
Change-Id: I82951719bfa8d069f56792cfd9caf3a27a9f6215
 2020-02-12 10:30:28 -08:00
Judd
d666b1bba5
Update wav_to_features.py 
Remove the leading comma in the generated float array. Withi this leading comma, the generated C code can't be compiled.
 2020-02-11 18:32:03 +08:00