Commit Graph

730 Commits (3bb22a6ff86d1b52d14b8d2ea826bd850bbe0137)

Author SHA1 Message Date
shugeo 02d8616692 Implementation of cuda kernel for fake_quant_with_min_max_vars_per_channels op. 2019-10-10 16:40:56 +03:00
shugeo 3504b0cda9 Implemented fake_quant_with_min_max_vars_per_channel fop cuda helper. The first working revision. 2019-10-10 15:44:50 +03:00
shugeo 753565145c Refactored fake_quant_with_min_max_vars op cuda implementation. 2019-10-10 14:00:49 +03:00
shugeo c13e945a96 Fixed fake_quant_with_min_max_vars op and tests. 2019-10-10 13:23:11 +03:00
Alexander Stoyakin fa8105bf0f
Merge pull request #9 from KonduitAI/asto_nd4s_indexing
[WIP] Indexing for INDArray in nd4s
2019-10-10 11:03:06 +03:00
Alexander Stoyakin 156fba4f77 Minor fix
Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-10-10 10:49:07 +03:00
Alexander Stoyakin be70bea359 Operator :: for INDArray
Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-10-10 10:43:32 +03:00
shugeo 3c0c59ab88 Refactored fake_quant_with_min_max_vars op. 2019-10-09 22:09:33 +03:00
shugeo 352f1eee80 Implemented fake_quant_with_min_max_per_channel helper for cpu platform. The first approach. 2019-10-09 21:39:59 +03:00
Alexander Stoyakin 0de8523ebd Indexing syntax changed
Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-10-09 16:07:20 +03:00
shugeo d0cbd33b0e Added input checks for op. 2019-10-09 15:52:13 +03:00
shugeo 3a89e51811 Added tests for fake_quant_with_min_max_vars_per_channel op. 2019-10-09 13:38:18 +03:00
Alexander Stoyakin f515151f5f
Merge pull request #8 from KonduitAI/asto_nd4s
[WIP] Indexing improvements in nd4s
2019-10-09 11:39:51 +03:00
Alexandre Boulanger 3aa51e210a RL4J: Extract TD Target calculations (StandardDQN and DoubleDQN) (#8267)
Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>
2019-10-09 09:14:47 +09:00
Alexander Stoyakin 8472782d7f More tests
Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-10-08 19:08:43 +03:00
shugeo cb56b0b06a The first approach for fake_quant_with_min_max_vars_per_channel op implementation. 2019-10-08 19:00:41 +03:00
Alexander Stoyakin 4956048a11 Tests added
Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-10-08 17:38:43 +03:00
Alexander Stoyakin b36b3fa1aa Some sugar
Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-10-08 16:41:12 +03:00
shugeo dbabd11f83
Merge pull request #7 from KonduitAI/shugeo_image_bordering
Shugeo image bordering
2019-10-08 15:47:37 +03:00
shugeo 8fe5a1fa96 The working implementation of draw_bounding_boxes op. 2019-10-08 15:42:27 +03:00
Robert Altena 50b13fadc8 nd4j-api cleanup. (#8273)
* nd4j-api cleanup.

Signed-off-by: Robert Altena <Rob@Ra-ai.com>

* restore deleted schemes.

Signed-off-by: Robert Altena <Rob@Ra-ai.com>
2019-10-08 21:48:22 +11:00
shugeo 30a8af566c The first working implementation of cuda kernel for draw_bounding_boxes op helper. 2019-10-08 13:45:18 +03:00
shugeo ae09cfee32 Next approach of cuda imlementation for draw_bounding_boxes op helper. 2019-10-08 00:09:46 +03:00
shugeo 6cf3a8fa9c Refactored cpu implementatio and added cuda aproach. 2019-10-07 17:51:07 +03:00
Alexander Stoyakin 7a1e327516 Indexing demo
Signed-off-by: Alexander Stoyakin <alexander.stoyakin@gmail.com>
2019-10-07 16:59:47 +03:00
shugeo 78443ffebf Working implementation of draw_bounding_boxes op for cpu. 2019-10-07 15:04:44 +03:00
shugeo 16a66a65e3 Added helper declaration for draw_bounding_boxes op. 2019-10-04 21:16:34 +03:00
shugeo 53a2ebddbe Added test and helpers for draw_bounding_boxes op both cpu and cuda related. 2019-10-04 20:46:26 +03:00
shugeo 8f70b4441f draw_bounding_boxes op implementation. Inital revision. 2019-10-04 18:32:21 +03:00
raver119 b8f2a83a5a
Merge pull request #5 from KonduitAI/shugeo_divnonan_full
Added implementation for divide_no_nan op and tests.
2019-10-04 10:33:04 +03:00
shugeo 908e4c4912 Added implementation for divide_no_nan op and tests. 2019-10-04 10:29:15 +03:00
raver119 2381c51480
Merge pull request #4 from KonduitAI/revert-3-shugeo_div_no_nan
Revert "Implement divide_no_nan op."
2019-10-03 20:26:15 +03:00
raver119 cff26f13c5
Revert "Implement divide_no_nan op." 2019-10-03 20:25:52 +03:00
raver119 bbfa41cbc8
Merge pull request #3 from KonduitAI/shugeo_div_no_nan
Implement divide_no_nan op.
2019-10-03 18:27:12 +03:00
shugeo 6eaca179d6 Implement divide_no_nan op. 2019-10-03 18:22:17 +03:00
raver119 d2e98564d4
Merge pull request #2 from KonduitAI/shugeo_bincast
[WIP] Shugeo bitcast
2019-10-03 14:27:24 +03:00
Robert Altena 1f4ad08305 refactoring activations. (#8261)
Signed-off-by: Robert Altena <Rob@Ra-ai.com>
2019-10-03 20:35:27 +10:00
shugeo 130ee25682 Implemented compare_and_bitpack op. 2019-10-03 10:57:48 +03:00
shugeo 75ad3c8153 Fixed test names. 2019-10-02 19:05:26 +03:00
shugeo f3e42173ef Refactored buffer copying to avoid wrong usage of buffers. 2019-10-02 16:51:09 +03:00
shugeo 1c6173d218 Added implementation of bitcast op. 2019-10-02 15:04:59 +03:00
shugeo a27e61553a Added tests and fixed op name. 2019-10-02 15:04:28 +03:00
shugeo 863ff76878 Added declaration for bincast op. 2019-10-02 12:17:00 +03:00
raver119 62326542e9
Merge pull request #1 from KonduitAI/shugeo_adjust
[WIP] Shugeo adjust
2019-10-01 14:15:54 +03:00
shugeo afeb524238 Refactored implementation for adjust_contrast ops. 2019-10-01 14:13:09 +03:00
shugeo 1575c704ae Added implementation for adjust_contrast_v2 op and tests. 2019-10-01 11:44:27 +03:00
raver119 44a8d19ac6
[WIP] Broadcast changes (#8257)
* - provide correct call NDArray::applyBroadcast inside of NDArray::applyTrueBroadcast

Signed-off-by: Yurii <yurii@skymind.io>

* - provide new trueBroadcast helper

Signed-off-by: Yurii <yurii@skymind.io>

* example for yurii

Signed-off-by: raver119 <raver119@gmail.com>

* - provide new trueBroadcast helper for cpu

Signed-off-by: Yurii <yurii@skymind.io>

* - start working on new trueBroadcat helper for cuda

Signed-off-by: Yurii <yurii@skymind.io>

* - further work on trueBroadcast for cuda

Signed-off-by: Yurii <yurii@skymind.io>

* - fix bugs in cuda helper trueBroadcast

Signed-off-by: Yurii <yurii@skymind.io>
2019-10-01 09:10:19 +03:00
Alexandre Boulanger 5959ff4795 RL4J: Fix QLearningDiscrete.setTarget() and add CartpoleNative (#8250)
* Fixed QLearningDiscrete.setTarget()

Signed-off-by: Alexandre Boulanger <aboulang2002@yahoo.com>

* Added native java version of Cartpole

Signed-off-by: unknown <aboulang2002@yahoo.com>
2019-10-01 09:27:51 +09:00
shugeo e06dfb5dcc Implementation of adjust_contrast op. 2019-09-30 18:24:12 +03:00
Alexandre Boulanger d5e98afcef RL4J: Add VideoRecorder (#8106)
* Added VideoRecorder

Signed-off-by: unknown <aboulang2002@yahoo.com>

* Added missing header

Signed-off-by: unknown <aboulang2002@yahoo.com>

* Changed HistoryProcessor to use VideoRecorder

Signed-off-by: unknown <aboulang2002@yahoo.com>
2019-09-30 13:40:32 +09:00