aa13eec | David Brownell | 16 July 2018, 21:57:16 UTC | Removed references to files that no longer exist. | 16 July 2018, 21:57:16 UTC |
d58886f | David Brownell | 16 July 2018, 21:52:50 UTC | Generated new code signing keys that expire on 12/31/2039 | 16 July 2018, 21:52:50 UTC |
9ce967a | Sergii Dymchenko | 09 July 2018, 22:24:50 UTC | Support one input case for Min/Max to pass ONNX backend tests. | 12 July 2018, 17:43:26 UTC |
dca867c | Bowen Bao | 10 July 2018, 23:07:06 UTC | add warning for convolution padding on channel axis. | 11 July 2018, 23:03:56 UTC |
5ca4bb3 | Bowen Bao | 16 June 2018, 01:07:43 UTC | Add Sequential Convolution. adding convolution over sequential axis related tests. adding convolution over sequential axis. currently additional supported parameters: auto padding strides groups support for dilation needs to be tested on GPU. updating PrimitiveOpType SerializationTests that is missing from other commits .. convert tabs to spaces. Refine cpp convolution unit tests. Add dilation tests to python convolution unit tests. more detailed comments on shape change for 1d seq conv with reduction rank 0. And other minor tweaks. add EndToEndTests of sequential convolution on MNIST add init_bias tests for seq conv minor change in comments rename ConvolutionOverSequenceAxisNode. Add comment on cudnn failed new test. add more comments, trim spaces add more comments, remove magic number, add more boundary checks. remove the last SetValue for outputSeqAxisDimValue as TensorView Unary Op has already updated the value. fix bug in python seqconv default bias shape, and add related unit tests. small tweak in seq conv to avoid additional gpu memory allocation and increase performance. Example: seq MNIST, and profiling adjust conv c++ value unit test channel size. small update on python seq mnist Sequential convolution v2. * re-designed ConvolutionSequenceShapeNode: refactored to separate out computing output sequence length from v1 node design. And refactored ConvolutionNodeBaseExtended as their common base class. (Since "ConvolutionNodeBase" is not only base class of ConvolutionNode but also PoolingNode). * Performance increase against v1. - compute sequence length by MBLayout instead of mask output from unpack. Avoiding the unnecessary cpu/gpu memory copy. not include py sequence example for now .. need to find they a correct location. add check for truncated sequences in sequential convolution improve code style. Moving sequential convolution in python to a new high level api, to maintain compatibility with previous implementation (special case 1d sequential convolution). Add ConvolutionSequenceShape OP. nit update conv_attribute test for updated convolution parameter move sequential parameter to the last update test shortcircuit for CPU convolution dilation. update endtoendtest - unittest baseline file for new convolution unittests. update makefile to include new unittest file for linux nit Update ConvolutionNode initialization code to handle TransformerNode Initialization. nit nit | 11 July 2018, 04:10:33 UTC |
8bd0459 | KeDengMS | 10 July 2018, 17:50:38 UTC | CPU device should not try to initialize nvml in NcclComm | 10 July 2018, 17:50:38 UTC |
5aa511d | David Brownell | 09 July 2018, 16:49:45 UTC | Project updates to use a consistent native runtime | 10 July 2018, 15:29:11 UTC |
1f0eed8 | Liqun Fu | 10 July 2018, 01:07:41 UTC | Merge branch 'liqun/gslfixstage' | 10 July 2018, 01:07:41 UTC |
83370c1 | TJ | 09 July 2018, 23:41:26 UTC | Squashed commit of the following: commit 02eb64bf5e9f6c22138b5111f5518f6087cea7e0 Author: TJ <tix@microsoft.com> Date: Mon Jul 9 13:07:13 2018 -0700 set the multiverso lib file when asgd is set to true, otherwise it will look for a lib that doesn't match any rule when asgd=false | 09 July 2018, 23:41:26 UTC |
3c87d20 | liqfu | 09 July 2018, 20:29:47 UTC | upgrade linux build from c++11 to c++14, enable gsl, update with latest LotusIR | 09 July 2018, 20:29:47 UTC |
55b4606 | Yang Chen | 09 July 2018, 17:01:33 UTC | Merge branch 'master' into yanchen/docker-copy | 09 July 2018, 17:01:33 UTC |
add2896 | David Brownell | 15 June 2018, 20:52:43 UTC | Not hard-coding build locations | 09 July 2018, 15:37:44 UTC |
e8f2db4 | Sergii Dymchenko | 07 July 2018, 00:01:07 UTC | Add bool in/out support to pass ONNX backend tests. | 07 July 2018, 04:50:53 UTC |
624bf7d | David Brownell | 07 July 2018, 01:47:29 UTC | String changes to support conversion from ASCII, UCS2, UCS4, UTF8, UTF16, and UTF32 strings | 07 July 2018, 01:48:27 UTC |
ed9de47 | Yang Chen | 07 July 2018, 00:54:06 UTC | Merge branch 'yanchen/mpi-makefile' | 07 July 2018, 00:54:06 UTC |
c11f51d | Sergii Dymchenko | 07 July 2018, 00:29:50 UTC | Merge branch 'sedymche/onnx-backend' | 07 July 2018, 00:29:50 UTC |
2626ae7 | Yang Chen | 06 July 2018, 21:36:58 UTC | Merge branch 'master' into yanchen/mpi-makefile | 06 July 2018, 21:36:58 UTC |
70cf3a3 | Yang Chen | 06 July 2018, 21:36:37 UTC | Merge branch 'master' into yanchen/docker-copy | 06 July 2018, 21:36:37 UTC |
7b3a840 | Sergii Dymchenko | 06 July 2018, 20:18:41 UTC | Update skip list for ONNX backend tests. | 06 July 2018, 20:20:39 UTC |
afbe91e | Bowen Bao | 06 July 2018, 18:08:36 UTC | update NoGPU.cpp | 06 July 2018, 18:08:36 UTC |
19ffa06 | Bowen Bao | 06 July 2018, 17:03:13 UTC | Merge branch 'master' into bowbao/gather_grad | 06 July 2018, 17:03:13 UTC |
6adce79 | Bowen Bao | 06 July 2018, 17:01:44 UTC | Merge branch 'master' into bowbao/gather_grad | 06 July 2018, 17:01:44 UTC |
10c1d6b | Bowen Bao | 06 July 2018, 16:54:26 UTC | nit | 06 July 2018, 16:54:26 UTC |
3bfa12b | TJ | 06 July 2018, 04:49:20 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into tix/fix_netcoreTestFailure | 06 July 2018, 04:49:20 UTC |
0b317eb | Bowen Bao | 06 July 2018, 02:04:49 UTC | Merge branch 'bowbao/clone_segfault' | 06 July 2018, 02:04:49 UTC |
92796cf | Bowen Bao | 06 July 2018, 01:37:40 UTC | fixed bug around mapping from columnValidMask to data. Instead of a 1 to 1 mapping, each element in columnValidMask maps to a grid of multiple rows and columns in data matrix. | 06 July 2018, 01:37:40 UTC |
04844b0 | TJ | 06 July 2018, 01:22:25 UTC | Fixed path | 06 July 2018, 01:22:25 UTC |
610abaf | TJ | 05 July 2018, 23:59:38 UTC | Added comment | 05 July 2018, 23:59:38 UTC |
a89a334 | TJ | 05 July 2018, 23:55:33 UTC | Fixed path | 05 July 2018, 23:55:33 UTC |
dc29eb3 | TJ | 05 July 2018, 23:52:10 UTC | Enabled the test for py35 so that it gets run by default in bvt | 05 July 2018, 23:52:10 UTC |
6cc772f | Bowen Bao | 05 July 2018, 23:42:51 UTC | small change for more reuse of code | 05 July 2018, 23:42:51 UTC |
56d0ada | TJ | 05 July 2018, 22:46:41 UTC | remove the data folder created by the python script | 05 July 2018, 22:46:41 UTC |
6e265da | Yang Chen | 05 July 2018, 21:33:54 UTC | Fixed an issue for copying patches Unlike Linux-like cp instruction, docker COPY instruction doesn't preserve the top-level dir structure (e.g. it copies the contents of the top-level dir rather than the dir itelse to the image). Our Dockerfile for generating GPU image has correct behavior though. | 05 July 2018, 21:42:02 UTC |
9a075b3 | Bowen Bao | 05 July 2018, 21:40:44 UTC | Merge branch 'master' into bowbao/clone_segfault | 05 July 2018, 21:40:44 UTC |
0df5c39 | Bowen Bao | 05 July 2018, 21:20:13 UTC | Add method ScatterToIndicesWithMask for class Matrix(both CPUMatrix and GPUMatrix). Change GatherNode Backprop to use the new method instead of altering data for gaps in input matrix. | 05 July 2018, 21:20:13 UTC |
49e3583 | Sergii Dymchenko | 04 July 2018, 00:42:01 UTC | Copy onnx_cntk from https://github.com/onnx/onnx-cntk. | 05 July 2018, 21:13:40 UTC |
9667426 | TJ | 05 July 2018, 19:12:36 UTC | Fixing run-test to grab for the correct dll for netcore build | 05 July 2018, 19:12:36 UTC |
ed9f879 | Yang Chen | 05 July 2018, 17:00:01 UTC | Merge branch 'master' into yanchen/mpi-makefile | 05 July 2018, 17:00:01 UTC |
17e6a01 | Sergii Dymchenko | 04 July 2018, 03:49:50 UTC | Merge branch 'pull/3277' | 04 July 2018, 03:49:50 UTC |
7c74244 | Bowen Bao | 04 July 2018, 01:48:01 UTC | set missing value back to 0 in case of unexpected usage. | 04 July 2018, 01:48:01 UTC |
1e058ce | Bowen Bao | 04 July 2018, 01:34:11 UTC | fix Gather op's incorrect gradient value. * the error was due to that we pad 0 as default value for missing gaps. All these then each contribute 1 to the gradient of reference at index 0. The fix is to mask missing values in indices matrix to negative, and in Matrix scatter implementation to check and skip negative indices. (previous Matrix CPU implementation already checks for negative indices) | 04 July 2018, 01:34:11 UTC |
c049493 | Yang Chen | 03 July 2018, 23:07:22 UTC | Merge branch 'master' into yanchen/mpi-makefile | 03 July 2018, 23:07:22 UTC |
81ced59 | Bowen Bao | 03 July 2018, 22:19:33 UTC | add None check for python clone substitution. * it seems None somehow escapes the typecheck when being converted to class CNTK::Variable thus causing the crash. | 03 July 2018, 22:19:33 UTC |
8577222 | TJ | 03 July 2018, 22:02:22 UTC | Merge branch 'tix/convertEvalExampleTestToNetCore' of https://github.com/Microsoft/CNTK | 03 July 2018, 22:02:22 UTC |
25a0896 | Yang Chen | 03 July 2018, 20:55:57 UTC | use system CXX for building Multiverso on Linux We used to set CXX to mpic++, which was then used for building Multiverso. Unfortunately, this kind of configuration conflicted with find_package(MPI), and hence caused failure. This commit fixed the issue by using system CXX for building Multiverso. It also fixed two other issues: * EVAL_LIB depends on libmultiverso, so we had to set MULTIVERSO_LIB before the EVAL_LIB rule * cmake variable doesn't have a "PATHNAME" type, we used PATH instead. We also replace FILEPATH with PATH because BOOST_LIBRARY_DIRS points to a path rather than a file. | 03 July 2018, 20:55:57 UTC |
dd9705b | Spandan Tiwari | 02 July 2018, 22:19:14 UTC | Fixing LRN op to match ONNX 1.2 spec. Size attribute has semantics of diameter and not radius, as was the case before. | 03 July 2018, 19:40:12 UTC |
8d39322 | TJ | 03 July 2018, 01:33:08 UTC | Converted eval example to use netcore | 03 July 2018, 01:33:08 UTC |
63603e0 | Peyman Manikashani | 02 July 2018, 18:39:26 UTC | SqueezeNet numbers mistmatch fix (due to an issue on softmax layer) | 02 July 2018, 20:58:49 UTC |
07fd96f | liqfu | 02 July 2018, 15:38:50 UTC | support ONNX1.2.2 cast op | 02 July 2018, 15:38:50 UTC |
2f91f74 | jediknight31 | 01 July 2018, 13:53:29 UTC | Fix saved model name output Corrected the output showing the name of the model being saved | 01 July 2018, 13:53:29 UTC |
bece037 | Sergii Dymchenko | 29 June 2018, 01:52:40 UTC | Merge branch 'sedymche/onnx-trig' | 29 June 2018, 01:52:40 UTC |
69938f2 | Spandan Tiwari | 28 June 2018, 22:34:30 UTC | Adding full support for 'alpha' attribute in ELU op in ONNX and CNTK. | 28 June 2018, 22:34:30 UTC |
e95f92c | Sergii Dymchenko | 27 June 2018, 23:11:29 UTC | Add Tan/Atan ops to CNTK (with ONNX support). | 28 June 2018, 21:03:02 UTC |
c589619 | liqfu | 28 June 2018, 15:38:36 UTC | update to ONNX1.2.2 | 28 June 2018, 15:38:36 UTC |
e160304 | Spandan Tiwari (CAFE) | 28 June 2018, 04:02:28 UTC | Merge branch 'sptiwari/onnx_avg_pool_stride' | 28 June 2018, 04:02:28 UTC |
6340792 | Jie Zhu | 28 June 2018, 00:55:55 UTC | Merge branch 'zhujie/straightthrough_unaryop' | 28 June 2018, 00:55:55 UTC |
2e29625 | TJ | 27 June 2018, 22:25:50 UTC | Merge branch 'tix_fixRestoredTestFailures' of https://github.com/Microsoft/CNTK | 27 June 2018, 22:25:50 UTC |
805c4a6 | Spandan Tiwari | 27 June 2018, 21:54:47 UTC | Fixing strides and dilation attributes in ONNX AveragePool, MaxPool, Conv, ConvTranpose import. | 27 June 2018, 21:54:47 UTC |
3c389ed | Jie Zhu | 08 June 2018, 16:38:32 UTC | fixing a bug in straight through implementation | 27 June 2018, 18:44:49 UTC |
cc368ae | Jie Zhu | 07 June 2018, 23:40:41 UTC | fixing a bug in backwarod propagation of straightthrough | 27 June 2018, 18:44:49 UTC |
8c1e5ed | Jie Zhu | 07 June 2018, 20:34:41 UTC | fixing type mismatch in tensorops.h and changing name of python name for straightthrough | 27 June 2018, 18:44:49 UTC |
02be8a0 | Jie Zhu | 06 June 2018, 22:55:37 UTC | adding straight through unary op | 27 June 2018, 18:44:49 UTC |
e612bc4 | Thiago Crepaldi | 26 June 2018, 23:58:24 UTC | Merge branch 'sptiwari/update_depth2space' | 26 June 2018, 23:58:24 UTC |
8a38f55 | TJ | 26 June 2018, 23:51:40 UTC | Changed run-test of CifarResNetClassifierTest to look for testData in the correct folder | 26 June 2018, 23:51:40 UTC |
da60e4f | Spandan Tiwari | 26 June 2018, 20:39:21 UTC | Updating DepthToSpace and SpaceToDepth ops to match ONNX spec. | 26 June 2018, 20:39:21 UTC |
ede74e8 | TJ | 26 June 2018, 17:08:21 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into tix_fixRestoredTestFailures | 26 June 2018, 17:08:21 UTC |
38dc974 | David Brownell | 24 June 2018, 19:50:45 UTC | Fixed behavior when dry_run is True | 26 June 2018, 15:13:26 UTC |
987b22a | Sergii Dymchenko | 23 June 2018, 01:18:32 UTC | Add Cos/Sin/Acos/Asin to ONNX export. | 26 June 2018, 00:03:31 UTC |
d19e4ca | Sergii Dymchenko | 20 June 2018, 18:10:10 UTC | Add sin/cos to ONNX import. | 25 June 2018, 23:59:26 UTC |
52101bf | TJ | 25 June 2018, 23:06:49 UTC | Copy v2libCStest dependencies to local | 25 June 2018, 23:06:49 UTC |
b7acd42 | TJ | 25 June 2018, 20:02:13 UTC | Added a flag to copy everything to output dir | 25 June 2018, 20:02:13 UTC |
c0b81b5 | David Brownell | 25 June 2018, 16:16:16 UTC | Restoring additional tests | 25 June 2018, 16:16:16 UTC |
ddea43c | David Brownell | 14 June 2018, 22:45:27 UTC | Removed code that was incorrectly checked in | 25 June 2018, 16:12:55 UTC |
29be2c8 | David Brownell | 14 June 2018, 21:36:11 UTC | Restored 1 more test | 25 June 2018, 16:12:55 UTC |
ffc4595 | David Brownell | 14 June 2018, 21:35:02 UTC | Restored tests | 25 June 2018, 16:12:55 UTC |
dc132a4 | David Brownell | 24 June 2018, 16:35:00 UTC | Fixed incorrect hash values | 24 June 2018, 16:35:00 UTC |
91afddb | Thiago Crepaldi | 23 June 2018, 20:38:19 UTC | Temporarily disable doctest until formatting is fixed | 23 June 2018, 20:38:19 UTC |
8f738ae | Thiago Crepaldi | 22 June 2018, 21:22:43 UTC | Merge branch 'master' into sptiwari/onnx1.2_merge | 22 June 2018, 21:22:43 UTC |
5df8651 | Thiago Crepaldi | 22 June 2018, 21:14:26 UTC | Increase error tolerance for BatchNormalization Spatial tests | 22 June 2018, 21:19:03 UTC |
e7976cc | Spandan Tiwari | 22 June 2018, 20:08:57 UTC | Merge branch 'master' into sptiwari/onnx1.2_merge | 22 June 2018, 20:08:57 UTC |
4cd3dc8 | David Brownell | 22 June 2018, 19:15:51 UTC | Disabling stack trace on Windows when environment var is set | 22 June 2018, 19:16:02 UTC |
117448f | Thiago Crepaldi | 21 June 2018, 22:39:15 UTC | Temporarily disable doctest until formatting is fixed | 22 June 2018, 17:26:14 UTC |
7ea5a2c | David Brownell | 22 June 2018, 01:15:13 UTC | Disabled additional C# test | 22 June 2018, 01:15:13 UTC |
623323f | Spandan Tiwari | 21 June 2018, 01:02:46 UTC | Temporarily disable op2cntk_test.py/{test_conv_setup,test_batch_norm_setup} to fix failures downstream | 21 June 2018, 22:34:55 UTC |
8f5fcd4 | Thiago Crepaldi | 21 June 2018, 17:23:31 UTC | Fix python doc tests due to formatting | 21 June 2018, 17:25:06 UTC |
134e976 | Jaliya Ekanayake | 19 June 2018, 21:55:04 UTC | fix style | 19 June 2018, 21:55:04 UTC |
79312de | Jaliya Ekanayake | 19 June 2018, 19:40:43 UTC | fix for the readerlib test failure. It was a quite a team work | 19 June 2018, 19:40:43 UTC |
7449613 | Peyman Manikashani | 19 June 2018, 17:20:00 UTC | Revert "Temporarily skip several failing float16 ONNX tests." This reverts commit deef207a6201015d2b7dbc6649a4e35cb47cfe8a. | 19 June 2018, 17:31:48 UTC |
6d85793 | Peyman Manikashani | 19 June 2018, 06:36:26 UTC | Adding support for loading fp16 data from TensorProto raw data field | 19 June 2018, 06:36:26 UTC |
2dc61c1 | Sergii Dymchenko | 18 June 2018, 21:01:41 UTC | Fix several float16 convolution ONNX tests on the CI system. | 18 June 2018, 21:04:28 UTC |
deef207 | Sergii Dymchenko | 16 June 2018, 02:25:12 UTC | Temporarily skip several failing float16 ONNX tests. | 16 June 2018, 02:25:12 UTC |
18910c5 | Sergii Dymchenko | 16 June 2018, 02:17:43 UTC | Skip test_OptimizedRNNStack. | 16 June 2018, 02:17:43 UTC |
11edc80 | Sergii Dymchenko | 16 June 2018, 02:16:08 UTC | Fix several float16 op tests on the build system. | 16 June 2018, 02:16:08 UTC |
172cab1 | Sergii Dymchenko | 16 June 2018, 01:06:01 UTC | Fix test_pooling and test_conv_with_freedim_model on the build system. | 16 June 2018, 01:55:52 UTC |
ea1010d | Sergii Dymchenko | 16 June 2018, 00:45:24 UTC | Fix test_convolution_transpose on the build system. | 16 June 2018, 01:55:50 UTC |
a59d5ee | liqfu | 16 June 2018, 01:20:31 UTC | Peyman GRU fp16 fix. Test fixes to disbale for fp16 conv. | 16 June 2018, 01:20:31 UTC |
58473d4 | liqfu | 15 June 2018, 21:47:22 UTC | Enable rnn ops with onnx 1.2 | 16 June 2018, 01:00:53 UTC |
c1a3db7 | Peyman Manikashani | 15 June 2018, 23:04:56 UTC | adding fp16 to all tests and disabling failing tests + updating schema.h to add fp16 | 15 June 2018, 23:47:14 UTC |
057c1ce | Thiago Crepaldi | 15 June 2018, 21:00:12 UTC | Fix cudnn symbolic link on CNTK Dockerfile | 15 June 2018, 21:05:38 UTC |
5371b53 | Spandan Tiwari | 15 June 2018, 21:04:00 UTC | Bringing back our fix for output node shape issue in Lotus graph.cpp. | 15 June 2018, 21:04:00 UTC |