601977b | liqun fu | 16 April 2020, 03:53:07 UTC | handle BN with sequence axis | 16 April 2020, 03:53:07 UTC |
e939648 | liqunfu | 31 March 2020, 15:55:14 UTC | disable fp16 test for poolWithSequenceAxis (#3810) | 31 March 2020, 15:55:14 UTC |
e1467a7 | liqunfu | 27 March 2020, 21:42:04 UTC | Merge pull request #3806 from microsoft/liqun/wrap_maxlpool_with_reshape support Pooling ops with Sequence axis | 27 March 2020, 21:42:04 UTC |
a2055f6 | liqun fu | 27 March 2020, 21:40:05 UTC | fix a typo | 27 March 2020, 21:40:05 UTC |
80e9f79 | liqun fu | 27 March 2020, 21:35:31 UTC | update with reviewers' comments | 27 March 2020, 21:35:31 UTC |
9a7dd4c | liqun fu | 25 March 2020, 19:37:46 UTC | add comments | 25 March 2020, 19:37:46 UTC |
280ec14 | liqun fu | 25 March 2020, 18:20:15 UTC | support Pooling ops with Sequence axis | 25 March 2020, 18:20:15 UTC |
c7bc93f | Thiago Crepaldi | 06 August 2019, 15:30:35 UTC | Fix Windows Test Dockerfile | 06 August 2019, 15:30:35 UTC |
9a4bbfe | Thiago Crepaldi | 31 July 2019, 17:49:03 UTC | Update Windows Test Dockerfile According to https://techcommunity.microsoft.com/t5/Containers/Windows-Server-2019-Now-Available/ba-p/382430, 'latest' tag is no longer supported. Using '1607' (windows server 2016) | 31 July 2019, 17:49:03 UTC |
bda8fcb | Yang Chen | 23 July 2019, 00:24:42 UTC | address CR | 23 July 2019, 00:24:42 UTC |
c3d7440 | Yang Chen | 22 July 2019, 23:48:15 UTC | add FutureValue Op's custom attributes into Scan node's description | 22 July 2019, 23:48:15 UTC |
91b78a7 | Yang Chen | 20 July 2019, 01:19:03 UTC | Merge branch 'yanchen/load-from-memory' Signed-off-by: Yang Chen <yanchen@microsoft.com> | 20 July 2019, 01:19:03 UTC |
f843f91 | Yang Chen | 19 July 2019, 02:23:45 UTC | Merge branch 'yanchen/rm-asserts' Signed-off-by: Yang Chen <yanchen@microsoft.com> | 19 July 2019, 02:23:45 UTC |
3ea8507 | Yang Chen | 18 July 2019, 22:11:00 UTC | Merge pull request #3722 from microsoft/yanchen/output-mapping added a mapping from onnx outputs to cntk outputs into the graph's description | 18 July 2019, 22:11:00 UTC |
d01ef8f | Yang Chen | 18 July 2019, 17:22:37 UTC | remove an assert that seems to be too conservative | 18 July 2019, 17:22:37 UTC |
cfe8f6c | Yang Chen | 18 July 2019, 07:52:21 UTC | Added a new API to load models from memory * added a new C api CNTK_LoadModel_FromArray that loads a model from memory * also enabled loading ONNX models from buffer | 18 July 2019, 07:57:45 UTC |
f7f1e8d | Yang Chen | 18 July 2019, 07:39:00 UTC | added a mapping from onnx outputs to cntk outputs into the graph's description | 18 July 2019, 07:39:00 UTC |
99a272b | KeDengMS | 05 July 2019, 23:18:23 UTC | Fix hard tabs | 05 July 2019, 23:18:23 UTC |
a4621a8 | KeDengMS | 05 July 2019, 20:33:45 UTC | Export custom attributes on Times node to ONNX file (#3709) | 05 July 2019, 20:33:45 UTC |
13f3b5f | liqfu | 27 June 2019, 17:26:40 UTC | Merge branch 'master' into liqun/scan14 | 27 June 2019, 17:26:40 UTC |
abf0985 | liqfu | 27 June 2019, 17:25:47 UTC | transpose batch with dir before reshape to [seq, batch, dir*hidden] for RNN ops | 27 June 2019, 17:25:47 UTC |
ac7a549 | Thiago Crepaldi | 25 June 2019, 19:16:56 UTC | Fix V2LibraryCSTests end to end test | 26 June 2019, 06:20:25 UTC |
9756a11 | Thiago Crepaldi | 25 June 2019, 19:16:56 UTC | Fix V2LibraryCSTests end to end test | 25 June 2019, 19:16:56 UTC |
ef67317 | liqfu | 21 June 2019, 21:50:18 UTC | set shape bacth dim to 0 for reshape op after LSTM to make batch size editable after ONNX export. | 21 June 2019, 21:50:18 UTC |
03a8c41 | liqfu | 19 June 2019, 16:24:52 UTC | default keep dim is true | 19 June 2019, 16:24:52 UTC |
1fe0bf1 | liqfu | 19 June 2019, 16:17:45 UTC | reduction keep dim edge cases | 19 June 2019, 16:17:45 UTC |
ad1709c | liqfu | 18 June 2019, 22:16:03 UTC | fix reduce op keepdim setting | 18 June 2019, 22:16:03 UTC |
248ff78 | liqfu | 11 May 2019, 00:03:27 UTC | onnx exporter for past/future value: fix for offset > sequence length cases; output delayed initial values | 11 May 2019, 00:03:27 UTC |
8ed7050 | liqfu | 01 May 2019, 21:06:26 UTC | ONNX convolution with more than 1 non-feature/channel dimensions | 01 May 2019, 21:06:26 UTC |
22a86bf | liqfu | 29 April 2019, 18:58:00 UTC | onnx export: fix cases when scan input needs to be broadcasted | 29 April 2019, 18:58:00 UTC |
0e172db | Thiago Crepaldi | 24 April 2019, 18:20:17 UTC | Fix build | 24 April 2019, 18:20:17 UTC |
82b7a39 | liqfu | 19 April 2019, 18:28:14 UTC | merge release 2.7 | 19 April 2019, 18:28:14 UTC |
6b72436 | Thiago Crepaldi | 09 April 2019, 19:54:42 UTC | Adds CNTK libs folder to LD_LIBRARY_PATH at CNTK init routine on Python | 09 April 2019, 19:54:42 UTC |
b8edaab | rpengms | 09 April 2019, 16:25:37 UTC | Rpengms/fp16 bs bugfix (#3632) Update CNTK.core.bs to add support for precision overwrite for every node | 09 April 2019, 16:25:37 UTC |
9688ac6 | rpengms | 01 April 2019, 17:14:06 UTC | fp16 - brain script add to CNTK.core.bs (#3617) | 01 April 2019, 17:14:06 UTC |
764c8c4 | Thiago Crepaldi | 25 March 2019, 18:08:19 UTC | Fix CNTK Tutorial 303 | 25 March 2019, 18:08:19 UTC |
49d4c2d | Thiago Crepaldi | 25 March 2019, 04:39:38 UTC | Update OOBE GPU Dockerfile to CUDA 10 | 25 March 2019, 04:39:38 UTC |
f48e11c | Thiago Crepaldi | 22 March 2019, 22:12:50 UTC | Update conda to >= 4.3.30 | 22 March 2019, 22:12:50 UTC |
3b7e3b9 | rpengms | 22 March 2019, 09:07:51 UTC | merge fp16_bs fix (#3609) fp16 brain script fix accumulator | 22 March 2019, 09:07:51 UTC |
4003c08 | rpengms | 20 March 2019, 18:36:16 UTC | Merge fp16 brainscript work (#3606) * FP16 BrainScript - address code review comments * Remove Tab and fix debug build breaks * Fix Linux Build breaks * fp16 brain script - add _CRT_SECURE_NO_WARNINGS * fp16 brain script - fix NetworkTests * Update tests for model version change * Remove changes for InputAndParamNodes * Fix typo * Remove redundant code * Fix optional parameters | 20 March 2019, 18:36:16 UTC |
45ae386 | liqfu | 17 March 2019, 21:36:13 UTC | disable Manual/Tutorial tests on Windows Python2.7 | 17 March 2019, 21:36:13 UTC |
9f60d8f | Peggy Peng | 13 March 2019, 17:07:22 UTC | Remove tab which caused build breaks | 15 March 2019, 17:47:16 UTC |
19ccd9f | Peggy Peng | 13 March 2019, 06:34:50 UTC | Address code review comments | 15 March 2019, 17:47:15 UTC |
3a21b0a | Peggy Peng | 15 February 2019, 23:11:23 UTC | FP16 all reduce. Cast gradients to and from float16 before sending over NCCL to do all reduce. | 15 March 2019, 17:47:14 UTC |
4765f69 | liqfu | 14 March 2019, 16:16:23 UTC | support onnx export of 2GB models | 14 March 2019, 16:16:23 UTC |
22ded46 | Peggy Peng | 28 February 2019, 22:42:37 UTC | Address code review comments | 28 February 2019, 22:43:45 UTC |
d46ad9d | Peggy Peng | 15 February 2019, 22:23:13 UTC | Update | 28 February 2019, 22:43:45 UTC |
773379e | rpengms | 11 February 2019, 22:07:54 UTC | Add more comments | 28 February 2019, 22:43:45 UTC |
d2dc123 | Peggy Peng | 04 December 2018, 19:21:28 UTC | Fix BMUF fp16 to update local learner's fp32 cache | 28 February 2019, 22:43:45 UTC |
26e26e4 | Peggy Peng | 30 November 2018, 20:42:30 UTC | Fix BMUF fp16 bug. Local learner ResetSmoothedGradients should NOT wipe cached fp32 model/parameters. | 28 February 2019, 22:43:44 UTC |
e104129 | Peggy Peng | 28 February 2019, 18:07:41 UTC | Remove obsolete comments | 28 February 2019, 18:09:44 UTC |
92054e8 | Peggy Peng | 15 February 2019, 22:19:07 UTC | Remove debug messages | 28 February 2019, 18:09:43 UTC |
730a220 | Peggy Peng | 13 February 2019, 19:32:28 UTC | Add useV2Aggregator to SGD config | 28 February 2019, 18:09:43 UTC |
6ef0224 | Thiago Crepaldi | 28 February 2019, 09:45:39 UTC | Merge branch 'pull/3431' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 28 February 2019, 09:45:39 UTC |
39ca0a2 | Thiago Crepaldi | 28 February 2019, 09:04:17 UTC | Merge branch 'pull/3482' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 28 February 2019, 09:04:17 UTC |
93f875c | Thiago Crepaldi | 28 February 2019, 06:49:34 UTC | Merge branch 'pull/3477' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 28 February 2019, 06:49:34 UTC |
8cbfbfe | Thiago Crepaldi | 28 February 2019, 04:38:53 UTC | Merge branch 'pull/3458' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 28 February 2019, 04:38:53 UTC |
ef5b745 | Thiago Crepaldi | 28 February 2019, 04:27:00 UTC | Merge branch 'pull/3455' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 28 February 2019, 04:27:00 UTC |
1c5f124 | Thiago Crepaldi | 28 February 2019, 02:24:07 UTC | Merge branch 'pull/3486' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 28 February 2019, 02:24:07 UTC |
726e48e | Thiago Crepaldi | 28 February 2019, 00:43:11 UTC | Merge branch 'pull/3491' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 28 February 2019, 00:43:11 UTC |
771d754 | Thiago Crepaldi | 27 February 2019, 23:59:43 UTC | Merge branch 'pull/3492' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 27 February 2019, 23:59:43 UTC |
5ccf54a | Thiago Crepaldi | 27 February 2019, 22:39:40 UTC | Merge branch 'pull/3514' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 27 February 2019, 22:39:40 UTC |
af42873 | Thiago Crepaldi | 26 February 2019, 21:07:30 UTC | Merge branch 'pull/3561' Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com> | 26 February 2019, 21:07:30 UTC |
06469f7 | Thiago Crepaldi | 25 February 2019, 18:46:26 UTC | Merge branch 'pull/3558' | 25 February 2019, 18:46:26 UTC |
a5fb3d6 | liqfu | 19 February 2019, 20:48:41 UTC | add more onnx rnn model tests, enable optimizedRNN op test | 19 February 2019, 20:48:41 UTC |
00b8037 | Bowen Bao | 16 January 2019, 23:21:08 UTC | Update onnx submodule, support ConstantOfShape & OneHot * Support SequenceIs[First,Last] with ConstantOfShape * Update bypass load test in verify\_one\_input & add test for one hot op * Update export for one hot op. Migrate from exporting onnx.ml.OneHotEncoder to onnx.OneHot. Op fixes * Fix topk onnx\_op\_test * Support MVN export using ONNX function * Fix LayerNormalization * Skip tests for sequence slice float16: not supported in cntk * Support gather export & import with float16 - Fix cntk gather issue with float16 inputs. - Support exporting constant float16 tensor. - Support importing int32 indices input for gather. * Enable more passed op tests | 15 February 2019, 23:12:10 UTC |
aa82819 | Liqun Fu | 14 February 2019, 21:07:34 UTC | Merge branch 'liqun/remove-crosstalkcaffe-test' | 14 February 2019, 21:07:34 UTC |
3bef0f4 | liqun fu | 13 February 2019, 19:08:06 UTC | protobuf 3.6.0 to 3.6.1 | 13 February 2019, 19:08:06 UTC |
9ce6c1c | Amit Agarwal | 08 February 2019, 16:37:36 UTC | Merge pull request #3550 from Microsoft/yaronwe/hide-protobuf hide protobuf library symbols | 08 February 2019, 16:37:36 UTC |
2ccdff4 | liqfu | 04 February 2019, 05:18:50 UTC | enable onnx rnn test, name input with proper name | 04 February 2019, 05:18:50 UTC |
979c8bf | Liqun Fu | 31 January 2019, 04:27:20 UTC | Merge branch 'liqun/liqun/scan6Stage2' | 31 January 2019, 04:27:20 UTC |
9093a4d | Thiago Crepaldi | 29 January 2019, 20:59:40 UTC | Cache CNTK dependencies at CNTK Azure Blob | 30 January 2019, 03:47:10 UTC |
6b2323c | liqfu | 28 January 2019, 22:48:13 UTC | onnx support of sequential convolution | 28 January 2019, 22:48:13 UTC |
631ef0f | Takashi Takebayashi | 28 January 2019, 15:02:51 UTC | Fix typo Ture -> True | 28 January 2019, 15:02:51 UTC |
4ef4417 | Peyman Manikashani | 18 December 2018, 23:50:23 UTC | Adding sequence models to CI and skipping failing ones | 24 January 2019, 23:01:14 UTC |
fc839bc | Peyman Manikashani | 23 January 2019, 00:18:07 UTC | Moving CI test dependencies to python environments yml files | 23 January 2019, 00:24:29 UTC |
d67a5ae | Aaron Schweiger | 16 January 2019, 22:00:33 UTC | Free nnzPerRowOrCol at end of SetValue | 16 January 2019, 22:00:33 UTC |
fea6e9d | Vadim Mazalov | 12 January 2019, 00:01:46 UTC | Merge branch 'vadimma/bmlfex' | 12 January 2019, 00:01:46 UTC |
62fa818 | Vadim Mazalov | 07 January 2019, 23:20:54 UTC | Enable frame mode for the binary MLF reader | 11 January 2019, 02:02:34 UTC |
0be4bcb | Liqun Fu | 09 January 2019, 02:39:18 UTC | Merge branch 'liqun/scan5Stage' | 09 January 2019, 02:39:18 UTC |
fe939aa | liqfu | 08 January 2019, 20:16:32 UTC | support ONNX scan(9) | 08 January 2019, 20:16:32 UTC |
a5fcc80 | Bowen Bao | 07 January 2019, 23:07:04 UTC | Resolve Linux binary drop & windows oobe test conda dependency | 07 January 2019, 23:07:04 UTC |
2274e02 | yaron weinsberg | 06 January 2019, 10:04:34 UTC | hide protobuf library symbols | 06 January 2019, 10:04:34 UTC |
b732aeb | Mickaël Schoentgen | 05 January 2019, 21:36:45 UTC | Fix several DeprecationWarning: invalid escape sequence Signed-off-by: Mickaël Schoentgen <contact@tiger-222.fr> | 05 January 2019, 21:36:45 UTC |
322293c | Vadim Mazalov | 04 January 2019, 00:21:27 UTC | Merge branch 'vadimma/adambs' | 04 January 2019, 00:21:27 UTC |
6c56fdd | liqfu | 03 January 2019, 23:45:28 UTC | Revert "refactor" This reverts commit 47345c6c54e95ed4a4948d5af9f1205f6e7c3932. | 03 January 2019, 23:45:28 UTC |
47345c6 | liqfu | 03 January 2019, 23:26:03 UTC | refactor | 03 January 2019, 23:26:03 UTC |
05435d0 | Vadim Mazalov | 18 December 2018, 23:36:45 UTC | Expose Adam in BS | 03 January 2019, 20:58:17 UTC |
08130db | Bowen Bao | 03 January 2019, 18:57:44 UTC | update linux binary drop cudnn path | 03 January 2019, 18:57:44 UTC |
e2d79d7 | Bowen Bao | 14 December 2018, 18:16:57 UTC | Submodule onnxruntime, and remove previous drop. * A few patches are required to build cntk_uwp. * Use proto from onnxruntime/protobuf instead of from onnx. * TODO: Some issues with onnx_op_test RNN and OptimizedRNNStack from shape inference. | 03 January 2019, 01:09:08 UTC |
254a336 | Bowen Bao | 21 December 2018, 18:40:18 UTC | Bump cuda version in nuget spec | 21 December 2018, 18:40:18 UTC |
bc70863 | Bowen Bao | 19 November 2018, 18:31:44 UTC | Add Maxunpooling export | 14 December 2018, 19:24:26 UTC |
d087acc | Bowen Bao | 13 December 2018, 18:37:03 UTC | Fix typo & update .yaml files * pyqt is required by matplotlib and various other libraries. In our .yaml files we default to install the latest version of pyqt. However the current latest version (5.9.2) is exhibiting some problems. Thus we downgrade to the stable version 5.6.0. | 13 December 2018, 18:37:03 UTC |
ef2f039 | Deyu Fu | 17 May 2018, 20:16:44 UTC | fix break for cub 1.8 API change | 13 December 2018, 00:14:15 UTC |
f178144 | Bowen Bao | 25 October 2018, 00:34:01 UTC | Support CUDA 10 * Move to support CUDA 10, cudnn 7.3, cub 1.8. * Fixed a bug related to "pointer to pin pointer is disallowed" #3063, which is exposed in newer version vctools. * Added workaround for a potential vs2017 15.9 bug with cntk Debug version. | 13 December 2018, 00:10:31 UTC |
93e1009 | liqfu | 12 December 2018, 03:17:16 UTC | comment debug code | 12 December 2018, 03:17:16 UTC |
de15bda | liqfu | 12 December 2018, 01:49:23 UTC | passed prod models and unfold test. handle delay op. | 12 December 2018, 01:49:23 UTC |
22e869e | Spandan Tiwari | 05 December 2018, 21:51:43 UTC | Add ONNX support for zeros_like, ones_like, and eye_like. | 05 December 2018, 21:51:43 UTC |
5fc6c2a | Peyman Manikashani | 30 November 2018, 19:22:25 UTC | sequence_isfirst_islast test fix for python 2.7 | 30 November 2018, 19:22:25 UTC |
75b0141 | Peyman Manikashani | 28 November 2018, 00:47:28 UTC | Adding support for exporting CNTK's Sequence::IsFirst and Sequence::IsLast nodes | 29 November 2018, 23:57:43 UTC |