4c0d724 | Eldar Akchurin | 04 October 2017, 12:25:30 UTC | Removing some prints | 04 October 2017, 12:25:30 UTC |
a856414 | Eldar Akchurin | 04 October 2017, 10:30:11 UTC | Fixing tabs | 04 October 2017, 10:30:11 UTC |
739d75f | Eldar Akchurin | 04 October 2017, 10:24:27 UTC | Fixing some tabs | 04 October 2017, 10:24:27 UTC |
4a587db | Eldar Akchurin | 04 October 2017, 10:16:23 UTC | Some fixes | 04 October 2017, 10:16:23 UTC |
2e54149 | Eldar Akchurin | 28 August 2017, 16:04:14 UTC | Fixing some stuff | 04 October 2017, 10:08:42 UTC |
e4d5dcb | Eldar Akchurin | 28 August 2017, 15:59:01 UTC | Exposing legacy sigmoid | 04 October 2017, 10:08:42 UTC |
8ffd19b | Eldar Akchurin | 25 August 2017, 17:41:13 UTC | Submitting Mudit's changes | 04 October 2017, 10:07:06 UTC |
0d13098 | Taehoon Lee | 04 October 2017, 05:00:27 UTC | Fix typos (#2434) | 04 October 2017, 05:00:27 UTC |
b224a7d | liqfu | 29 September 2017, 02:01:29 UTC | fix rest issues find with code scan | 29 September 2017, 02:01:29 UTC |
af149ee | Eldar Akchurin | 25 September 2017, 12:52:27 UTC | Refactoring of general graph algorithms from ComputationNetwork | 28 September 2017, 10:32:57 UTC |
7243389 | hecli | 28 September 2017, 05:16:26 UTC | Merge branch 'master' into liqun/ArgFix | 28 September 2017, 05:16:26 UTC |
3da3f5a | liqfu | 27 September 2017, 21:25:49 UTC | extra va_end_start pair | 27 September 2017, 21:25:49 UTC |
bded222 | Yuqing Tang | 27 September 2017, 16:59:17 UTC | Corrected docstring of Adam optimizer. | 27 September 2017, 16:59:17 UTC |
fceab10 | Project Philly | 27 September 2017, 09:12:56 UTC | Integrate juncgu/optimizingAgg into master | 27 September 2017, 09:12:56 UTC |
f43c6bf | Project Philly | 27 September 2017, 08:13:19 UTC | Integrate kedeng/issue2388 into master | 27 September 2017, 08:13:19 UTC |
79e0039 | Junjie Qian | 22 August 2017, 22:02:37 UTC | Fix to build on CPU fix build on Windows | 27 September 2017, 02:46:18 UTC |
afc8deb | Juncheng Gu | 02 August 2017, 19:47:38 UTC | optimize AggregateGradientsImpl (async pipeline) for non-GDR & GPU scenario rename "j" to currentGradientIndex in line321 update description in line309; rename j to currentGradientIndex fix bug (from Junjie) in new gradient aggregator pipeline (== -> !=) move first D-to-H copy out of the loop define "size_t i" in simple gradient aggregator | 27 September 2017, 02:46:16 UTC |
62a80d2 | KeDengMS | 27 September 2017, 00:06:33 UTC | Fix Issue #2388 | 27 September 2017, 00:06:33 UTC |
40162ba | Nikos Karampatziakis | 26 September 2017, 18:46:50 UTC | faster way to get the logits in the nce docstring | 26 September 2017, 18:46:50 UTC |
6120812 | Cha Zhang | 25 September 2017, 22:55:13 UTC | Update README.md | 25 September 2017, 22:55:13 UTC |
a6302a9 | liqfu | 25 September 2017, 22:01:57 UTC | push | 25 September 2017, 22:01:57 UTC |
dfef3d6 | Ivan Rodriguez | 11 September 2017, 14:06:09 UTC | Avoiding segmentation fault while introducing udf in recurrent loop | 25 September 2017, 11:59:11 UTC |
474d468 | motamed | 24 September 2017, 12:36:25 UTC | minor bug fix (#2402) | 24 September 2017, 12:36:25 UTC |
1cbccd0 | Emad Barsoum | 21 September 2017, 20:03:28 UTC | Merge branch 'master' into ebarsoum/dilation_with_padding Merge with master... | 21 September 2017, 20:03:28 UTC |
99d2c82 | KeDengMS | 21 September 2017, 06:14:16 UTC | Pick up submodule fixes for block momentum in python and add test Remove some dead code | 21 September 2017, 06:14:16 UTC |
3bf78e8 | Emad Barsoum | 21 September 2017, 00:12:38 UTC | Fix dilation convolution bug with auto padding | 21 September 2017, 00:12:38 UTC |
ec8d6d6 | Vadim Mazalov | 11 September 2017, 05:37:31 UTC | Update the CTC tutorial further | 19 September 2017, 19:15:40 UTC |
7fb2acf | Vadim Mazalov | 31 August 2017, 18:14:50 UTC | update CTC tutorial | 19 September 2017, 19:15:40 UTC |
0c6ebba | Spandan Tiwari | 19 September 2017, 00:14:47 UTC | Adding group convolution. | 19 September 2017, 00:14:47 UTC |
c3e4f66 | Mark Hillebrand | 13 September 2017, 14:20:33 UTC | Tests/Install/linux/prep-run-test.sh: do not run CNTK_104_Finance_Timeseries_Basic_with_Pandas_Numpy.ipynb | 18 September 2017, 17:59:05 UTC |
96cdd12 | hecli | 12 September 2017, 20:08:40 UTC | Bump version numbers | 18 September 2017, 17:58:47 UTC |
bced9cf | Zhou Wang | 29 August 2017, 12:44:47 UTC | remove inclusion of dependency dlls, as it has been referenced in UWP nuget package. | 18 September 2017, 17:58:31 UTC |
e4cc788 | Mark Hillebrand | 13 September 2017, 09:07:46 UTC | Tests/EndToEndTests: Linux: test green for Tesla M60 | 18 September 2017, 09:04:32 UTC |
83dab7c | KGWANGMIN | 18 September 2017, 05:11:59 UTC | Fix typos (#2370) This PR fixes some typos: CNKT | 18 September 2017, 05:11:59 UTC |
75fa318 | Taehoon Lee | 17 September 2017, 19:38:46 UTC | Fix typos (#2212) | 17 September 2017, 19:38:46 UTC |
cd6748f | Joannes Vermorel | 17 September 2017, 19:37:41 UTC | Handfew typos fixed (#2366) | 17 September 2017, 19:37:41 UTC |
4a5f64c | KGWANGMIN | 17 September 2017, 01:08:08 UTC | Fix typos (#2338) This PR fixes some typos: CTNK | 17 September 2017, 01:08:08 UTC |
159a611 | Darío Hereñú | 17 September 2017, 01:07:39 UTC | Minor fixes (proposal) (#2357) plus typo? | 17 September 2017, 01:07:39 UTC |
6cc4451 | Reza Farrahi M | 17 September 2017, 01:06:26 UTC | Fixes "'exp' was not declared" (#2364) This fixes the error: "error: ‘exp’ was not declared in this scope return momTC == 0.0 ? 0 : exp(-1.0 / momTC);",when compiling CNTK with GCC. Reference: https://github.com/Microsoft/CNTK/issues/2341 | 17 September 2017, 01:06:26 UTC |
f275a54 | Cha Zhang | 16 September 2017, 12:11:58 UTC | Update README.md | 16 September 2017, 12:11:58 UTC |
2f9e469 | REDMOND\sayanpa | 15 September 2017, 22:50:17 UTC | Added sequence classification to current language understanding tutorial | 15 September 2017, 22:50:17 UTC |
bfe1d3c | Mark Hillebrand | 12 September 2017, 13:36:27 UTC | Tests/EndToEndTests: add some CC5 baselines for Linux | 15 September 2017, 10:54:23 UTC |
741d01e | Daniel Schneider | 15 September 2017, 06:14:51 UTC | language edits | 15 September 2017, 06:14:51 UTC |
ef49c98 | Project Philly | 15 September 2017, 06:08:38 UTC | Integrate yuqtang/StageManualUseLearners2 into master | 15 September 2017, 06:08:38 UTC |
c525d71 | Project Philly | 15 September 2017, 00:58:42 UTC | Integrate yuqtang/StageCleanStartupPrints2 into master | 15 September 2017, 00:58:42 UTC |
26fa020 | Yuqing Tang | 14 September 2017, 23:23:44 UTC | Squashed commit of the following: Updated the manual on how to use learners to explain the new Learner APIs for release 2.2. | 14 September 2017, 23:23:44 UTC |
0f3033d | Project Philly | 14 September 2017, 22:55:00 UTC | Integrate yuqtang/StagingFixNewLearner2 into master | 14 September 2017, 22:55:00 UTC |
d49774d | Yuqing Tang | 14 September 2017, 22:19:29 UTC | Corrected the case of the word "version". | 14 September 2017, 22:19:29 UTC |
8c87d1a | Cha Zhang | 14 September 2017, 22:19:14 UTC | Update release notes. | 14 September 2017, 22:19:14 UTC |
1e32914 | Yuqing Tang | 14 September 2017, 22:12:17 UTC | Squashed commit of the following: Removed sensitive information in CNTK start-up message. | 14 September 2017, 22:12:17 UTC |
a01f8cd | Yuqing Tang | 14 September 2017, 20:01:55 UTC | Squashed commit of the following: Add python side work-around to allow the combination of mean gradient and specifying number samples with the right implementation; in Python, achieve the legacy logic of having per sample learning rate with mean gradient by overriding the per sample learning rate with ignored_minibatch_size. Also clarification in docstring. | 14 September 2017, 20:01:55 UTC |
e48892c | duli1 | 14 September 2017, 18:11:34 UTC | Merge branch 'master' of https://github.com/Microsoft/cntk | 14 September 2017, 18:11:34 UTC |
0eb7a1f | duli1 | 14 September 2017, 18:10:55 UTC | update keras and tensorboard text | 14 September 2017, 18:10:55 UTC |
d9b0da2 | Project Philly | 14 September 2017, 06:08:59 UTC | Integrate liqun/iterationnote2 into master | 14 September 2017, 06:08:59 UTC |
484a4f7 | KeDengMS | 14 September 2017, 05:58:15 UTC | style fix | 14 September 2017, 05:58:15 UTC |
10fb523 | liqun fu | 14 September 2017, 05:42:12 UTC | CSharp iteration note | 14 September 2017, 05:42:12 UTC |
f27ed0a | duli1 | 14 September 2017, 05:10:27 UTC | Merge branch 'master' of https://github.com/Microsoft/cntk | 14 September 2017, 05:10:27 UTC |
b4a1704 | duli1 | 14 September 2017, 05:08:47 UTC | update contributor list. | 14 September 2017, 05:08:47 UTC |
035d79f | Project Philly | 14 September 2017, 02:03:38 UTC | Integrate ebarsoum/iteration into master | 14 September 2017, 02:03:38 UTC |
9ccbc67 | duli1 | 14 September 2017, 01:34:28 UTC | Merge branch 'master' of https://github.com/Microsoft/cntk | 14 September 2017, 01:34:28 UTC |
66b996f | duli1 | 14 September 2017, 01:31:04 UTC | update with keras and opencv info | 14 September 2017, 01:34:20 UTC |
d007293 | Emad Barsoum | 14 September 2017, 01:29:10 UTC | Merge with master. | 14 September 2017, 01:29:10 UTC |
5d9369e | hecli | 14 September 2017, 01:21:00 UTC | format change | 14 September 2017, 01:21:00 UTC |
e31a11a | hecli | 14 September 2017, 01:13:07 UTC | Add acknowledgments | 14 September 2017, 01:13:07 UTC |
e16816e | KeDengMS | 14 September 2017, 01:07:38 UTC | More style update | 14 September 2017, 01:07:38 UTC |
99adb85 | KeDengMS | 14 September 2017, 01:01:06 UTC | Fix style in current_iteration.md | 14 September 2017, 01:01:06 UTC |
d5cf5e1 | KeDengMS | 14 September 2017, 00:57:02 UTC | current_iteration.md update on NCCL2, NVidia FP16 and Intel MKL works | 14 September 2017, 00:57:02 UTC |
7dd2ce1 | Emad Barsoum | 14 September 2017, 00:34:15 UTC | Update iteration planning. | 14 September 2017, 00:34:15 UTC |
4dfd1e2 | Zhou Wang | 13 September 2017, 22:28:50 UTC | fix version | 13 September 2017, 22:28:50 UTC |
78f122c | Zhou Wang | 13 September 2017, 22:17:19 UTC | add more text | 13 September 2017, 22:17:50 UTC |
adccabc | Daniel Schneider | 13 September 2017, 21:58:56 UTC | Update LICENSE.md changed license verbiage for Intel MKL to allow us to move up to MKL 2017 | 13 September 2017, 21:58:56 UTC |
6c2f2a1 | Project Philly | 13 September 2017, 21:34:33 UTC | Integrate zhouwang/releasenote-update into master | 13 September 2017, 21:34:33 UTC |
6f5f285 | Spandan Tiwari | 13 September 2017, 21:23:36 UTC | Adding links to object detection tutorial/demos in 2.2 release notes. | 13 September 2017, 21:23:36 UTC |
7ca42da | Zhou Wang | 13 September 2017, 21:19:13 UTC | add reference to doc page | 13 September 2017, 21:19:13 UTC |
162c7c5 | Spandan Tiwari | 13 September 2017, 21:10:33 UTC | Removing semantic segmentation from 2.2 release note. | 13 September 2017, 21:10:33 UTC |
4a875a1 | Spandan Tiwari | 13 September 2017, 21:07:25 UTC | Updated FasterRCNN and convolution FreeDimension bullets in 2.2 release note. | 13 September 2017, 21:07:25 UTC |
1797ff3 | Project Philly | 13 September 2017, 20:53:37 UTC | Integrate sptiwari/fastrcnn_readme into master | 13 September 2017, 20:53:37 UTC |
bcfbfbc | sayanpa | 13 September 2017, 20:46:17 UTC | Added links to the manuals as well. | 13 September 2017, 20:46:17 UTC |
0e24a13 | sayanpa | 13 September 2017, 20:40:51 UTC | Added link to python doc site. | 13 September 2017, 20:40:51 UTC |
50100fe | Spandan Tiwari | 13 September 2017, 20:34:20 UTC | Minor updates to FastRCNN readme files. | 13 September 2017, 20:34:20 UTC |
039f3f5 | Yuqing Tang | 13 September 2017, 03:32:10 UTC | Squashed commit of the following: Updated the new learner API description in the release notes. | 13 September 2017, 03:32:10 UTC |
3dfbebb | Spandan Tiwari | 13 September 2017, 00:46:15 UTC | Adding release note for free static axes support for convolution. | 13 September 2017, 00:46:15 UTC |
3fa4e36 | liqun fu | 12 September 2017, 18:31:34 UTC | correct CSharp name, disable a flaky test for further investigation | 12 September 2017, 18:31:34 UTC |
f004aea | Zhou Wang | 12 September 2017, 16:46:33 UTC | add information about examples to relesae notes | 12 September 2017, 16:59:18 UTC |
bfce957 | Project Philly | 12 September 2017, 15:54:59 UTC | Integrate liqun/LogisticRegFix into master | 12 September 2017, 15:54:59 UTC |
351e702 | Mark Hillebrand | 12 September 2017, 11:18:10 UTC | Tests/Install/linux/prep-run-test.sh: bump Jupyter TO to 2700 | 12 September 2017, 14:57:58 UTC |
e5b4bdc | liqun fu | 12 September 2017, 14:43:36 UTC | quick fix logistic regression build break | 12 September 2017, 14:43:36 UTC |
c76b6f0 | Mark Hillebrand | 12 September 2017, 12:58:09 UTC | Tests/EndToEndTests/TestDriver.py: add major CC for Tesla M60 | 12 September 2017, 12:58:09 UTC |
f89f9d7 | Mark Hillebrand | 12 September 2017, 08:45:21 UTC | Scripts/install/linux/install-cntk.sh: install libtiff5 | 12 September 2017, 10:17:45 UTC |
ab8f4e5 | Project Philly | 12 September 2017, 08:10:48 UTC | Integrate liqun/MakeDrop2.2Stage into master | 12 September 2017, 08:10:48 UTC |
0f4044e | Project Philly | 12 September 2017, 07:27:40 UTC | Integrate nikosk/attention4 into master | 12 September 2017, 07:27:40 UTC |
f41a1d2 | Project Philly | 12 September 2017, 05:44:31 UTC | Integrate sptiwari/remove_installfastrcnn into master | 12 September 2017, 05:44:31 UTC |
16d01d7 | Project Philly | 12 September 2017, 05:35:08 UTC | Integrate yuqtang/StagingRefactorLearningCore into master | 12 September 2017, 05:35:08 UTC |
738b740 | Nikos Karampatziakis | 12 September 2017, 02:43:46 UTC | Attention overhaul | 12 September 2017, 02:43:46 UTC |
10a68a4 | Project Philly | 12 September 2017, 01:57:48 UTC | Integrate liqun/CSApiRegressionExample into master | 12 September 2017, 01:57:48 UTC |
2b3e104 | Project Philly | 12 September 2017, 01:50:31 UTC | Integrate DaveyBiggers-master into master | 12 September 2017, 01:50:31 UTC |
9c34c47 | Project Philly | 12 September 2017, 01:45:07 UTC | Integrate liqun/CSApiRegressionExampleStage into master | 12 September 2017, 01:45:07 UTC |
ccfd6c7 | Yuqing Tang | 12 September 2017, 00:36:39 UTC | Refactor the core learner APIs: * Simplified TrainingParameterSchedule * Added minibatch_size to TrainingParameterSchedule and Learner * Added minibatch size to Learner * Added learning_parameter_schedule to Python APIs * Marked as deprecated on UnitType, learning_parameter_schedule, momentum_as_time_constant_schedule * Added minibatch_size parameter to all the learners: sgd, momentum_sgd, adam and all others except for user defined learner * Simplified learner parameter logic: If minibatch_size == cntk.learners.IGNORE, act as the parameters are applied applied to the whole minibatch; otherwise, scaling the parameters according to the actual minibatch_size and the minibatch_size specified during the the schedules and the learners construction time. More subtle details are in the docstrings and the comments. | 12 September 2017, 00:36:39 UTC |
e70a023 | Project Philly | 11 September 2017, 23:33:03 UTC | Integrate chenta/free_dimension_pooling into master | 11 September 2017, 23:33:03 UTC |
a3f20af | liqfu | 11 September 2017, 19:51:47 UTC | Add PretrainedModels to windows/linux/examples drops | 11 September 2017, 19:51:47 UTC |