https://github.com/Microsoft/CNTK

sort by:
Revision Author Date Message Commit Date
9758c3b Fix version incompatibility 09 May 2018, 18:57:59 UTC
51ed19a Integrate db211f1d84965fd56f450cd4592f0c99b3b1d99c into master 27 April 2018, 01:46:53 UTC
de5621c Fixing Convolution layer doc to include dilation and groups arguments. 25 April 2018, 22:30:51 UTC
adf4826 Merge pull request #3156 from MalKeshar/patch-1 Allow to add extra cxxflags via env 25 April 2018, 17:59:22 UTC
98dc8e4 Integrate gloch/fix_native_proposallayerlib into master 24 April 2018, 21:23:28 UTC
d5db843 Add Linux OOBE support 24 April 2018, 20:35:46 UTC
575e21b Fix VS build when source code and VS are in different drives 24 April 2018, 20:09:24 UTC
d1e2c91 Fix native ProposalLayerLib dimensions indexes 24 April 2018, 14:16:20 UTC
e785a2b Allow to add some extra cxxflags via env Allow to add some extra cxxflags via system env. Use cases: 1) https://github.com/Microsoft/CNTK/issues/3155 export CXXFLAGS="-Wno-unused-variable -Wno-sign-compare" configure && make 2) Our libc/libstdc has been built with gcc-4.9. So we fail to build cntk in current state with gcc-5 in compatibility mode. https://gcc.gnu.org/onlinedocs/libstdc++/manual/using_dual_abi.html After patching cntk we have successful build: COMMON_FLAGS="-D_GLIBCXX_USE_CXX11_ABI=0" CXXFLAGS="-D_GLIBCXX_USE_CXX11_ABI=0 -Wno-unused-variable -Wno-sign-compare" configure && make (nvcc don't recongnise -W flags) 24 April 2018, 13:36:56 UTC
51b9c0e Fix conv2d bug when having conv2d over the static axes of sequences. 23 April 2018, 22:55:46 UTC
438f619 Merge pull request #3140 from Joe8Bit/master (chore) Cleanup ./README.md 23 April 2018, 17:35:44 UTC
cd03a71 Update README.md 23 April 2018, 17:35:27 UTC
086073e adding publication date 23 April 2018, 17:31:29 UTC
5a0f072 adding related works reference to DSSM tutorial 23 April 2018, 17:29:47 UTC
116b662 Add Windows OOBE support 23 April 2018, 17:11:52 UTC
ab0ef38 Update dependencies from VS2015 to VS2017 and Anaconda3 4.1.1 to 4.3.1 23 April 2018, 17:11:52 UTC
1964906 Fix PythonBindings build on machines with VS2017 23 April 2018, 17:11:52 UTC
69a95f8 Integrate liqun/RNNBlk2 into master 23 April 2018, 17:01:48 UTC
401e511 Use TLS 1.2 in devInstall script. 20 April 2018, 00:09:26 UTC
baf3c66 change cprintf to use vsnprintf (#3124) change cprintf to use vsnprintf 19 April 2018, 20:05:55 UTC
0b78d90 (chore) Cleanup ./README.md * Reorganise the sections to better reflect the the main use cases of the `READMD.md` e.g. finding out about the project in general, rather than a changelog * Add more and cleanup markdown format (e.g. consistent single backticks for inline code, and python syntax highlighting for code blocks) * Gramatical/syntax fixes 19 April 2018, 16:46:11 UTC
133d346 Update documentation for 2.5.1 17 April 2018, 23:10:34 UTC
e237559 Bump release version to 2.5.1 17 April 2018, 23:06:45 UTC
78a5ff3 Add support to build wheel packages with third party libs 17 April 2018, 23:02:21 UTC
0b9b253 Added group convolution to current_iteration .md 17 April 2018, 20:34:14 UTC
db211f1 Remove unnecessary sys/io.h Machines with aarch64 architecure such as NVIDIA Jetson TX1 does not have sys/io.h. Without this commit, building libCntk.Core-2.5.so fails on a aarch64 machine. It should be safe to completely remove the inclusion of sys/io.h, since it seems that functions in the header are not used by CNTK. 16 April 2018, 01:16:59 UTC
1ffdc6e Integrate sptiwari/shufflenet into master 14 April 2018, 05:53:15 UTC
cf839dc Fixing ONNX BatchNormalization implementation to match spec. 13 April 2018, 23:14:34 UTC
2e7bbaa Fix windows script to install cntk from binary drop 13 April 2018, 22:11:48 UTC
cd9285c make GRU/RNNStep blocks, issue 3084 to expose BlockArgumentsMapping 13 April 2018, 21:12:45 UTC
89db005 Integrate kedeng/RNNfp16_2 into master 12 April 2018, 20:23:41 UTC
5e0856e Fix bugs for fp16 in RNN and BMUF Note that sparse embedding is still not working yet. 11 April 2018, 22:35:08 UTC
8bd55b3 Donot abort early if some ranks failed to initial NCCL 11 April 2018, 22:06:56 UTC
09e25a4 Moving group convolution implementation to use cuDNN7 and MKL2017 APIs. 11 April 2018, 17:26:15 UTC
6e7ff0e Add requirements check for CNTK during Python module import 10 April 2018, 00:16:59 UTC
d3d26d0 Integrate v-rodemo/fix-batchnorm-freedimension into master 09 April 2018, 23:39:20 UTC
82c8c69 Fixing the reader lib bug when number of samples in a chunk is less than the number of workers 08 April 2018, 06:18:46 UTC
d4781c1 updating script to use new learning APIs 06 April 2018, 19:05:45 UTC
0b5b100 Added test for batch norm spatial shape inference 06 April 2018, 16:19:39 UTC
ca0f889 Integrate yuqtang/StageGradOnGatherIndices into master 04 April 2018, 22:30:13 UTC
455aef8 Integrate vadimma/fb into master 04 April 2018, 07:48:31 UTC
0ebaed4 Add an assert for defining serializer 03 April 2018, 20:35:13 UTC
5afcff3 Fix definesMbSize config in bundler 28 March 2018, 22:16:16 UTC
90a41f6 Enabled gather op indcies be computed from parameters. 27 March 2018, 23:48:11 UTC
a7453ed Merge pull request #3065 from Microsoft/szhao/fixsample Update cntk version to 2.5 in the sample 26 March 2018, 22:06:07 UTC
09cdc6c fix the cntk version in the sample 24 March 2018, 10:45:46 UTC
e6cb4c5 Fix spatial batchnorm parameter inference with FreeDimension 23 March 2018, 19:57:45 UTC
ed4bd3b Integrate jaliyaek/recur_checkin into master 23 March 2018, 01:33:54 UTC
a5e0912 Fix automated cntk install script for windows 23 March 2018, 00:16:50 UTC
5b04f46 Integrate sptiwari/add_rnn_to_ornn2 into master 22 March 2018, 23:24:40 UTC
73c2046 Adding recurrence support to user defined functions. This enables UDF to be called inside recurrent loops. 22 March 2018, 18:44:09 UTC
3f72469 Adding ONNX support for rnnReLu and rnnTanh in OptimizedRNNStack. 22 March 2018, 18:24:32 UTC
65961c9 Integrate liqun/2954 into master 21 March 2018, 22:44:06 UTC
82142f0 Integrate muditj/FixNoRandomizerBug3 into master 21 March 2018, 21:19:55 UTC
aab2567 Integrate liqun/RNNStage into master 21 March 2018, 19:23:28 UTC
c8c02bf Fix no randomizer bug where workers see duplicate data 21 March 2018, 18:57:19 UTC
3e83c56 issue 2954: c# API memory leak 21 March 2018, 18:14:28 UTC
08ba969 support ONNX RNN 21 March 2018, 16:57:54 UTC
3a22962 Integrate sptiwari/mean_var_norm_issue into master 21 March 2018, 05:23:48 UTC
92b7b27 Update .clang-format to remove column limits Otherwise too much code change when formatting with clang-format 21 March 2018, 01:57:47 UTC
7068329 Fixes to mean_variance_normalization and ONNX LSTM to run Office models. 20 March 2018, 17:53:06 UTC
10c327e No aggregation or NCCL initialization if only one node used undef _SCL_SECURE_NO_WARNINGS for training 19 March 2018, 16:54:41 UTC
542406d No need to aggregate with one worker in brainscript Remove msg printint GPU UUID 19 March 2018, 16:50:09 UTC
91dbed4 Integrate vadimma/readse into master 17 March 2018, 20:32:53 UTC
6942ad1 Update README.md with correct links to examples 17 March 2018, 02:59:16 UTC
7e1a653 Update README and .clangformat 17 March 2018, 00:46:46 UTC
296ac76 Preparing Release 2.5 - update documentation 16 March 2018, 22:21:50 UTC
34ed567 Preparing Release 2.5 - bump version number 16 March 2018, 22:21:38 UTC
e39aeae Integrate liqun/it20183 into master 15 March 2018, 17:03:41 UTC
0f428b3 iteration update 15 March 2018, 15:47:22 UTC
0b79085 Fixing ONNX producer_version bug. 15 March 2018, 07:21:47 UTC
4178d9f Integrate liqun/ONNXIRupdateStage2 into master 15 March 2018, 04:09:25 UTC
7fd48e7 Updating release notes with ONNX and other feature work. 15 March 2018, 00:22:54 UTC
a895b76 update with ONNXIR 14 March 2018, 22:28:08 UTC
3a1b8e1 update with ONNXIR 14 March 2018, 21:47:58 UTC
18723c7 Integrate liqun/GRUStage into master 14 March 2018, 19:00:05 UTC
6608da7 Add logs for reader debugging 14 March 2018, 18:25:20 UTC
46b6016 Add ONNX GRU support 14 March 2018, 16:30:30 UTC
feb00c1 Integrate yuqtang/ValidateRecurerenceStepFunction into master 14 March 2018, 02:30:11 UTC
e61a7f0 Adding distributed GAN example 13 March 2018, 23:51:06 UTC
edbda9d Added sanitary check for step function signature with appropriate error messages. 13 March 2018, 21:38:57 UTC
f1f3bb4 Support ONNX GRU 13 March 2018, 18:24:33 UTC
44c626a Fix wheel package name for nightly build scripts WHL package name has recently changed for GPU target build. Now the package is called cntk-gpu, which translates to a filename prefixed as cntk_gpu*.whl 13 March 2018, 17:03:28 UTC
b3daad9 Add 'long_description' for wheel package's metadata file 13 March 2018, 17:02:26 UTC
769ea5c Adding OptimizedRNNStack ONNX support. Also Identity and NoOp support. 13 March 2018, 04:06:04 UTC
4b5c085 Fix the default device issue in test 12 March 2018, 15:26:56 UTC
82b56d1 Fixing the distributed learner sample count issue 12 March 2018, 08:00:49 UTC
ae81129 Integrate jqian/nccl into master 10 March 2018, 03:27:56 UTC
3d0af05 Integrate kedeng/fixCSBaseline into master 10 March 2018, 02:16:33 UTC
00c7944 Fix OneHotSkip for C# 32-bit SizeTVector 09 March 2018, 23:37:05 UTC
405f4be Integrate kedeng/fix3004B into master 09 March 2018, 23:27:46 UTC
966f7ad Update pip package metadata (setup.py) 09 March 2018, 22:00:55 UTC
d63ddd9 Fix Issue #3004 for mpi=no 09 March 2018, 19:48:53 UTC
487a8bb Update baseline following commit 678b8a9 09 March 2018, 19:18:59 UTC
b709af3 Integrate kedeng/fix3019B into master 09 March 2018, 08:32:30 UTC
3cb3430 Integrate 678b8a9a986b4423449fca4ecdd2a492438dcc28 into master 09 March 2018, 01:45:36 UTC
f6b3260 Fix issue #3019: RuntimeError: OptimizedRNNStackNode configured for sequence mode, but minibatch only has one time step. 09 March 2018, 00:37:09 UTC
28f587f Get GPU ID with nvmlDeviceGetUUID instead of device id in /proc/ Thanks to Frank's initial code change. Initialize NCCL with GPU ID frm nvml call instead of from device id in /proc. Previous way causes "same device being used" error if two different containers running on same host and try to initialize NCCL. Add nvml library for nvml apis Update the Makefile 08 March 2018, 18:41:37 UTC
dd88885 Fall back to MPI if failed to initialize NCCL 08 March 2018, 18:41:37 UTC
4f03bad Fix memory leak in MKL layout 08 March 2018, 07:11:22 UTC
back to top