https://github.com/Microsoft/CNTK

sort by:
Revision Author Date Message Commit Date
ed5451c MakeBinaryDrop: Fix lib names 01 February 2018, 03:32:04 UTC
c528a56 Update README.md 31 January 2018, 20:35:40 UTC
490a58e Add Halide to Release notes 31 January 2018, 20:08:57 UTC
2196299 MakeBinaryDrop: Change Cuda libs to 9.0 31 January 2018, 19:58:40 UTC
3df48dd Add release notes 31 January 2018, 18:31:11 UTC
1afe404 Replace 2.4+ with 2.4 31 January 2018, 14:05:05 UTC
1f15d46 Bump version to 2.4 31 January 2018, 14:04:24 UTC
b09ff2b Merge branch 'master' into liqun/onnx17Stage2 31 January 2018, 13:50:43 UTC
4bb3031 Making axis attribute optional in Concat. 31 January 2018, 08:52:52 UTC
22cea7c broadcast of scalar 31 January 2018, 04:48:15 UTC
e91df39 Pooling missing code fix 30 January 2018, 22:14:36 UTC
f18bcda Updating release notes with ONNX and other feature work (e.g. group convolution). 30 January 2018, 18:05:47 UTC
2a764a7 add ImageScaler, fix ConvTranspose 30 January 2018, 16:46:06 UTC
00158ff Merge pull request #2887 from Microsoft/marhamil-fix-jar Update post-build.cmd 30 January 2018, 03:13:01 UTC
6ea2c67 Break Nuget packages Separate out dependencies from Nuget packages to reduce size of each individual package because Nuget service do not allow to upload a package of size more than 333MB. Current list of packages: CNTK.Deps.Cuda - Cuda dlls for GPU package CNTK.Deps.MKL - MKL dlls CNTK.Deps.OpenCV.Zip - OpenCV and Zip dlls CNTK.CPUOnly - CNTK CPU only Nuget package. Dependencies: MKL, OpenCV.Zip CNTK.GPU - CNTK GPU Nuget package. Dependencies: Cuda, MKL, OpenCV.Zip CNTK.UWP.CPUOnly - CNTK CPU Nuget package for UWP. Microsoft.Research.CNTK.CpuEval-mkl - Legacy CNTK eval Nuget package. Dependencies: MKL 30 January 2018, 02:35:28 UTC
6e846aa fix makefile 29 January 2018, 20:44:29 UTC
8569868 fix variable 29 January 2018, 18:53:32 UTC
166c8bc add debug dll support 29 January 2018, 16:12:37 UTC
6db8b6e Add image writer SOs and zlib dlls 29 January 2018, 15:58:34 UTC
08cc45c Edit makefile 29 January 2018, 15:56:46 UTC
be931c8 Update post-build.cmd 29 January 2018, 15:56:45 UTC
ba9c2e7 Fix build if optional MKLDNN is not present 28 January 2018, 19:27:32 UTC
950ac47 Integrate jaliyaek/config into master 28 January 2018, 12:25:33 UTC
dd47e21 Making Halide default in Jenkin builds 27 January 2018, 01:17:58 UTC
343f383 Remove Python 3.4 support 26 January 2018, 22:55:22 UTC
ffc7507 Integrate jaliyaek/halide_squash2 into master 25 January 2018, 07:14:14 UTC
a7a52d7 Adding halide based binary convolution operators and its dependancies 25 January 2018, 00:41:17 UTC
851ea5d Fix linux binary drop script for nccl 24 January 2018, 06:18:32 UTC
387725d Merge branch 'kedeng/fixDocker2' 24 January 2018, 03:28:47 UTC
9f6cced Fix Dockerfiles under Tools/docker 23 January 2018, 21:44:01 UTC
a087490 Integrate sptiwari/remove_find_by_uid into master 23 January 2018, 20:16:49 UTC
a40285c Rename default Anaconda python environment Depending on the CNTK setup steps followed, either cntk-py<version> or <cntkdev-py<version> is installed. cntk-py<version> will be adopted as the default from now on. 23 January 2018, 18:43:51 UTC
e883520 Remove redefinition of CNTK_COMPONENT_VERSION Math.vcsproj inherits this macro from CNTK.Common.props already 23 January 2018, 18:28:53 UTC
3cf3af5 CNTK support for CUDA 9 CNTK now supports CUDA 9/cuDNN 7. This requires an update to build environment to Ubuntu 16/GCC 5 for Linux, and Visual Studio 2017/VCTools 14.11 for Windows. With CUDA 9, CNTK also added a preview for 16-bit floating point (a.k.a FP16) computation. Please check out the example of FP16 in ResNet50 at /Examples/Image/Classification/ResNet/Python/TrainResNet_ImageNet_Distributed.py Notes on FP16 preview: * FP16 implementation on CPU is not optimized, and it's not supposed to be used in CPU inference directly. User needs to convert the model to 32-bit floating point before running on CPU. * Loss/Criterion for FP16 training needs to be 32bit for accumulation without overflow, using cast function. Please check the example above. * Readers do not have FP16 output unless using numpy to feed data, cast from FP32 to FP16 is needed. Please check the example above. * FP16 gradient aggregation is currently only implemented on GPU using NCCL2. Distributed training with FP16 with MPI is not supported. * FP16 math is a subset of current FP32 implementation. Some model may get Feature Not Implemented exception using FP16. * FP16 is currently not supported in BrainScript. Please use Python for FP16. To setup build and runtime environment on Windows: * Install [Visual Studio 2017](https://www.visualstudio.com/downloads/) with following workloads and components. From command line (use Community version installer as example): vs_community.exe --add Microsoft.VisualStudio.Workload.NativeDesktop --add Microsoft.VisualStudio.Workload.ManagedDesktop --add Microsoft.VisualStudio.Workload.Universal --add Microsoft.Component.PythonTools --add Microsoft.VisualStudio.Component.VC.Tools.14.11 * Install [NVidia CUDA 9](https://developer.nvidia.com/cuda-90-download-archive?target_os=Windows&target_arch=x86_64) * From PowerShell, run: /Tools/devInstall/Windows/DevInstall.ps1 * Start VCTools 14.11 command line, run: cmd /k "%VS2017INSTALLDIR%\VC\Auxiliary\Build\vcvarsall.bat" x64 --vcvars_ver=14.11 * Open /CNTK.sln from the VCTools 14.11 command line. Note that starting CNTK.sln other than VCTools 14.11 command line, would causes CUDA 9 [build error](https://developercommunity.visualstudio.com/content/problem/163758/vs-2017-155-doesnt-support-cuda-9.html). To setup build and runtime environment on Linux using docker, please build Unbuntu 16.04 docker image using Dockerfiles /Tools/docker. For other Linux systems, please refer to the Dockerfiles to setup dependent libraries for CNTK. 23 January 2018, 00:58:56 UTC
eeb0645 Removing find_by_uid API. 22 January 2018, 18:33:41 UTC
3765da9 Support CNTK installation in VirtualEnv CNTK installation using pip place dlls in the root folder of the python environment. Unlike Conda, VirtualEnv doesn't add this folder in library search path and hence 'import CNTK' is broken in VirtualEnv. Fix: Place CNTK libraries in site-packages/CNTK/lib and add the same to library search path. 20 January 2018, 01:59:03 UTC
997d3a1 Replace CntkCsAssemblyVersion by CntkComponentVersion on VS projects 19 January 2018, 22:36:34 UTC
f36c73c Integrate sptiwari/find_by_uid4 into master 19 January 2018, 19:41:04 UTC
4836687 Fixing bug in ONNX pooling node serialization. 17 January 2018, 23:10:23 UTC
f07a98d Adding find_by_uid to logging API. 17 January 2018, 22:35:23 UTC
8066621 Fixing BPTT for case when the minibatch size changes 17 January 2018, 01:20:23 UTC
c3d01e4 Enabling evaluation on GPU 17 January 2018, 01:20:23 UTC
5bdaed7 Integrate yuqtang/TimesOnFreeAxes into master 16 January 2018, 23:27:02 UTC
10d7130 Integrate vadimma/LatticeSerializer into master 16 January 2018, 22:39:23 UTC
8b6af80 Update the max_sequence_length 16 January 2018, 03:41:23 UTC
05cbd0a Remove too much logging 16 January 2018, 00:40:06 UTC
e6dc1a4 Refactor lattice sequence initialization' 15 January 2018, 23:22:11 UTC
77c95a4 Trim lattice TOC line entries 15 January 2018, 08:08:02 UTC
d929ecd Add logs for lattice index builder 15 January 2018, 07:45:18 UTC
1a81d41 Fix batch matmul test failures 15 January 2018, 07:32:12 UTC
d26f547 debug large lattice entries 15 January 2018, 04:57:11 UTC
1fe17f8 Add some debugging statements 15 January 2018, 01:13:02 UTC
f79324c Ensure the lattice serializer handles several chunks in a file 15 January 2018, 00:17:15 UTC
ea6834b Restrict the size of lattice 14 January 2018, 17:57:29 UTC
5962f71 Introduce the maxSequenceSize config to the HTK feature reader 14 January 2018, 06:54:37 UTC
7d67daf Enable max seq size config in HTK reader 14 January 2018, 00:16:40 UTC
15e705d add batch matmul 13 January 2018, 02:41:34 UTC
01c22ad Change the docstring 13 January 2018, 01:38:02 UTC
bcbf95b Some fixes after rebase 12 January 2018, 20:07:47 UTC
f3e9993 Remove main from the lattice test 12 January 2018, 19:06:41 UTC
2cac661 Update the python test 12 January 2018, 19:06:41 UTC
44f21e7 Update python test 12 January 2018, 19:06:40 UTC
eb88155 Include the python test and bump up the model version 12 January 2018, 19:06:40 UTC
0a6a429 Update baselines to ensure they match 12 January 2018, 19:03:24 UTC
9c70a43 Python lattice reader test 12 January 2018, 19:03:23 UTC
5bb82fa Expose lattice reader to python 12 January 2018, 19:03:23 UTC
697793b Ensure the e2e tests between old and new readers match 12 January 2018, 19:03:23 UTC
bf673b8 Disable parallel lattice constructoin 12 January 2018, 19:03:23 UTC
25d79f1 Update the SE with new reader test 12 January 2018, 19:03:23 UTC
73f2315 Ensure the clean up builds 12 January 2018, 19:03:22 UTC
e57bac1 Further lattice reader cleanup 12 January 2018, 19:03:22 UTC
5489a4e Change the old lattice reader config back 12 January 2018, 18:59:43 UTC
9f1d5e2 Update old lattice reader test to 1 utterances 12 January 2018, 18:59:43 UTC
aaaa939 Expose latticedeserializer in python 12 January 2018, 18:59:43 UTC
ce3ff8c Remove baselines for update 12 January 2018, 18:59:42 UTC
75f3e43 Ensure lattice sequence owns the buffer 12 January 2018, 18:59:42 UTC
6f9185f Revert changes of the HTKMLFreader 12 January 2018, 18:59:42 UTC
e230130 Some clean of the lattice reader change 12 January 2018, 18:59:42 UTC
8c60dca Ensure the new reader output is identical to the old reader 12 January 2018, 18:59:42 UTC
1d04199 Forcibly moving allocation of lattices to CPU 12 January 2018, 18:59:41 UTC
733a4e7 Fixing mbsize 12 January 2018, 18:59:41 UTC
6e39b34 Clean up boundary code 12 January 2018, 18:59:41 UTC
0ea1eaf Remove current dir logging 12 January 2018, 18:59:41 UTC
d4e71b8 Return the SequenceTraining baselines 12 January 2018, 18:59:41 UTC
0adc7be Update e2e test baselines 12 January 2018, 18:59:41 UTC
a5b78ee Some more debugging 12 January 2018, 18:59:40 UTC
8912cbf Some debugging of failed linux test 12 January 2018, 18:59:40 UTC
fcb3a2f Add some logs for the lattice reader' 12 January 2018, 18:59:40 UTC
4135324 Add check if lattice index file exists 12 January 2018, 18:59:40 UTC
bd783aa Debug statement for the Linux test 12 January 2018, 18:59:40 UTC
c848303 Update E2E test for lattice reader 12 January 2018, 18:59:40 UTC
646cf60 Revert pre-compute break statement 12 January 2018, 18:59:39 UTC
13206a4 Add lattice index builder to makefile 12 January 2018, 18:59:39 UTC
4b975a0 Explicitly reference SequenceWithSoftmax node methods 12 January 2018, 18:59:39 UTC
554610e Some gcc fixes 12 January 2018, 18:59:39 UTC
9565deb Remove tabs 12 January 2018, 18:59:39 UTC
d38c060 Add new lattice reader test 12 January 2018, 18:59:39 UTC
60c2996 Some optimization of the binary reader 12 January 2018, 18:59:38 UTC
0750688 some debugging statements 12 January 2018, 18:59:38 UTC
36315e7 Some asserts and prints 12 January 2018, 18:59:38 UTC
back to top