Skip to main content

Browse the archive

https://github.com/Microsoft/CNTK
03 November 2020, 07:30:59 UTC
sort by:
Revision Author Date Message Commit Date
cf7fc9f Alexey Kamenev 18 December 2015, 23:17:49 UTC Updated baselines for cuDNN. 18 December 2015, 23:42:35 UTC
b2fb4d0 Frank Seide 18 December 2015, 23:00:05 UTC added plumbing for tensor operations to GPUMatrix.cu, but have not implemented an acty; implemented row-major conversion in GPUMatrix::SetValue() (on the CPU, so not maximally efficient) 18 December 2015, 23:00:05 UTC
3a6ea21 Frank Seide 18 December 2015, 21:31:26 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/tensors 18 December 2015, 21:31:26 UTC
c343e98 Frank Seide 18 December 2015, 21:30:33 UTC further optimized the most frequent tensor loops (1-stride loops for unary and binary ops), but still not seeing 4-way SSE parallelism 18 December 2015, 21:30:33 UTC
9d33fc1 Frank Seide 18 December 2015, 18:55:30 UTC added a specialization of a tensor op for inner dimensions where all strides are 1. Seems not quite enough for really efficient unrolling though 18 December 2015, 18:55:30 UTC
1a1bd17 Frank Seide 18 December 2015, 18:01:17 UTC bug fix: ComputationNode::DetermineNumCols() was an outdated pre-refactoring hold-over with a now incorrect validity check. Can just be removed, Should fix reported by user xiaoqing; removed unnecessary ad inconsistent use of 'this->' throughout Matrix.cpp, also fixed some bad indentations 18 December 2015, 18:01:17 UTC
1f26215 Alexey Kamenev 17 December 2015, 20:31:17 UTC Fixed mbStart in ImageReader for distributed case. 18 December 2015, 17:29:40 UTC
b8de2fe Alexey Kamenev 17 December 2015, 02:52:17 UTC Added support for distributed reading in ImageReader. 18 December 2015, 17:29:30 UTC
6e20025 Frank Seide 18 December 2015, 16:57:49 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/tensors 18 December 2015, 16:57:49 UTC
91eadb0 Frank Seide 18 December 2015, 16:54:19 UTC moved all tensor ops to a new header TensorOps.h so they can be shared between matrix types; also moved the float/double-unified math overloads (e.g. exp_()) there, as well as additional typically needed functions such as Sigmoid() 18 December 2015, 16:54:19 UTC
679c3c5 Mark Hillebrand 15 December 2015, 12:39:43 UTC Source/Readers/LMSequenceReader/: also build SequenceWriter on Linux 18 December 2015, 11:59:40 UTC
12e312f Frank Seide 18 December 2015, 08:17:45 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/tensors 18 December 2015, 08:17:45 UTC
f54e1fe Frank Seide 18 December 2015, 08:07:59 UTC implemented unary and ternary tensor ops. CPU implementation of elementwise tensor ops is feature complete (but may require optimization) 18 December 2015, 08:07:59 UTC
7d32cdf Frank Seide 18 December 2015, 06:41:19 UTC implemented all binary tensor operators (don't we love macros!) 18 December 2015, 06:41:19 UTC
83e5bbc Qiwei Ye 18 December 2015, 04:38:39 UTC Revert "Revert "adding an MPI init test in case of that MPI was initialized repeatedly"" This reverts commit 23ebe452a5e35dddfba2d08e8fb3265901bfc8af. 18 December 2015, 04:38:39 UTC
928da88 Frank Seide 18 December 2015, 01:04:55 UTC first version of CPU implementation of TensorView::DoSumOf() working now 18 December 2015, 01:04:55 UTC
38cb2fa Frank Seide 18 December 2015, 00:14:54 UTC bug fix in MBLayout: We should not guard against all parallel sequences having a gap at a time step, as that happens in truncated BPTT, and it would be much more complex to fix the reader, so we allow it 18 December 2015, 00:14:54 UTC
aa5d1a7 Frank Seide 17 December 2015, 23:50:00 UTC implemented plumbing and first shot for TensorView operation with reduction 17 December 2015, 23:50:00 UTC
8588af4 Frank Seide 17 December 2015, 19:35:54 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/tensors 17 December 2015, 19:35:54 UTC
e6040d0 Frank Seide 17 December 2015, 19:35:28 UTC made Linux build happy (missing explicit method template specialization of CPUMatrix<char>::Resize()) 17 December 2015, 19:35:28 UTC
bb6fc1b Frank Seide 17 December 2015, 19:33:52 UTC optimized MBLayout::InitAsFrameMode(), short-replacing calls to AddSequence() by a much simpler direct initialization for this special case; added editing functions to TensorShape, and rewrote TensorView::DoBinaryOpOf() to use them 17 December 2015, 19:33:52 UTC
cd6543e yzhang87 17 December 2015, 18:55:41 UTC A quick fix to the Kaldi Reader (sequence training, need more test!!) 17 December 2015, 18:55:41 UTC
23ebe45 Qiwei Ye 17 December 2015, 02:44:33 UTC Revert "adding an MPI init test in case of that MPI was initialized repeatedly" This reverts commit f55e6e4e694204a19861268ee0b1e4201fcba366. 17 December 2015, 02:44:33 UTC
397eac2 Frank Seide 17 December 2015, 00:02:18 UTC first experimental prototype of elementwise tensor op in PlusNode::ForwardProp() done except for actual kernel; new methods TensorShape::Pad() and Concat(); new method ComputationNode::GetTensorsForwardBinary(); moved ElementWiseOperator to CommonMatrix.h, using it in TensorView::DoSumOf(); TensorView::m_sob changed from ref to pointer to make the object copyable 17 December 2015, 00:02:18 UTC
583f10e Frank Seide 16 December 2015, 21:59:18 UTC fixed up where some header files showed up in the CNTK Project after the directory move 16 December 2015, 21:59:18 UTC
eb14fb7 Frank Seide 16 December 2015, 21:53:30 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/tensors 16 December 2015, 21:53:30 UTC
ceef627 Frank Seide 16 December 2015, 21:52:59 UTC made gcc happy (AsVector<>()) 16 December 2015, 21:52:59 UTC
4d96c29 Frank Seide 16 December 2015, 21:48:15 UTC added a constructor of TensorShape from IConfigRecord; minor simplifications in ScriptableObjects 16 December 2015, 21:48:15 UTC
3ab0dda thhoens 16 December 2015, 17:53:10 UTC Fix for LibSVMBinaryReader to add prefetching, microbatches, and Linux support. 16 December 2015, 20:13:38 UTC
132d72c thhoens 16 December 2015, 17:52:37 UTC Changed MemAllocator to use void* pointers instead of char*, and gave CUDAPageLockedMemAllocator static methods. 16 December 2015, 20:13:38 UTC
c30c9cd Jasha Droppo 16 December 2015, 18:09:25 UTC Merge branch 'master' into CUDA-elementwise-rework 16 December 2015, 18:09:25 UTC
bd4273e Frank Seide 16 December 2015, 18:08:09 UTC adapted README with pathnames of VS Debug command lines to new directory structure 16 December 2015, 18:08:09 UTC
bbcd0e5 Mark Hillebrand 16 December 2015, 15:51:19 UTC Cosmetic changes in .filters files 16 December 2015, 15:51:19 UTC
bff4577 Mark Hillebrand 16 December 2015, 15:39:54 UTC Move KaldiReaderReadme. Remove some non-referenced, obsolete, or empty files. 16 December 2015, 15:39:54 UTC
9d06597 Mark Hillebrand 16 December 2015, 15:39:01 UTC CNTK.sln: break out Other folder into top-level items, add Tools This makes Documentation more visible. 16 December 2015, 15:39:01 UTC
c9a343b Philipp Kranen 16 December 2015, 14:55:05 UTC Fixed paths for ParallelTraining 16 December 2015, 14:55:05 UTC
4ae7f76 pkranen 16 December 2015, 14:37:58 UTC Fixed build-and-test script for Example move 16 December 2015, 14:37:58 UTC
9ec5b2c pkranen 16 December 2015, 14:31:00 UTC Moved Examples for directory resturcturing 16 December 2015, 14:34:09 UTC
a9c4922 Mark Hillebrand 16 December 2015, 14:24:49 UTC Makefile: adapt to new location of generate_build_infop script 16 December 2015, 14:24:49 UTC
bfe490e Mark Hillebrand 16 December 2015, 13:54:33 UTC Add execute permissions for Python and shell scripts 16 December 2015, 13:54:33 UTC
4a888ad pkranen 16 December 2015, 12:58:53 UTC Renamed Scripts flder to Tools 16 December 2015, 12:58:53 UTC
8cb85a5 pkranen 16 December 2015, 11:57:49 UTC Moved EndToEndTests 16 December 2015, 11:57:49 UTC
33bb02c Philipp Kranen 16 December 2015, 08:42:02 UTC Fixed vcxproj.filters wrt restructuring 16 December 2015, 08:42:02 UTC
0f926d9 Philipp Kranen 16 December 2015, 08:21:53 UTC Fixed scattered test project files 16 December 2015, 08:21:53 UTC
caf67c7 Mark Hillebrand 15 December 2015, 08:58:24 UTC Adapt Linux build 16 December 2015, 08:05:32 UTC
6595b06 pkranen 16 December 2015, 07:15:30 UTC Moved TensorViw files according to new structure 16 December 2015, 07:35:14 UTC
f355292 pkranen 15 December 2015, 20:38:58 UTC Moved tests for directory restructuring 16 December 2015, 07:35:13 UTC
c3895bb pkranen 15 December 2015, 19:45:10 UTC Move Source files for directory restructuring and fix project dependencies 16 December 2015, 07:35:11 UTC
1b3aead Frank Seide 16 December 2015, 05:20:28 UTC undid an accidental change of _assignToRowSliceValuesOf() call, discovered by Yu Zhang 16 December 2015, 05:20:28 UTC
8e1e392 Frank Seide 16 December 2015, 03:43:33 UTC made PCH happy (added stdafx.h) 16 December 2015, 03:43:33 UTC
14bb711 Frank Seide 16 December 2015, 03:33:34 UTC made gcc happy (printf format) 16 December 2015, 03:33:34 UTC
a8c9c73 Frank Seide 16 December 2015, 03:31:04 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/tensors 16 December 2015, 03:31:04 UTC
f94b1da Frank Seide 16 December 2015, 03:30:40 UTC (some steps forward in TensorView) 16 December 2015, 03:30:40 UTC
4022de2 Alexey Kamenev 16 December 2015, 01:37:52 UTC Fix for convo unit tests when cuDNN is not supported. 16 December 2015, 01:37:52 UTC
4b9c82a Frank Seide 16 December 2015, 01:23:00 UTC towards a prototype of a binary op 16 December 2015, 01:23:00 UTC
94e49de Frank Seide 16 December 2015, 00:31:59 UTC VS forgot to add the new files to the git repo... 16 December 2015, 00:31:59 UTC
7fc742e Frank Seide 15 December 2015, 23:19:16 UTC Merge branch 'fseide/mblayout' of https://git.codeplex.com/cntk into fseide/tensors 15 December 2015, 23:19:16 UTC
1286da7 Frank Seide 15 December 2015, 23:07:17 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/mblayout 15 December 2015, 23:07:17 UTC
54b0e0c Frank Seide 15 December 2015, 23:03:06 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/tensors 15 December 2015, 23:03:06 UTC
aec884a Frank Seide 15 December 2015, 23:02:43 UTC started the TensorView class by adding header and CPP to math lib 15 December 2015, 23:02:43 UTC
3a0ec41 Alexey Kamenev 14 December 2015, 23:40:39 UTC Enabled convo unit tests. Added batch normalization sample. 15 December 2015, 20:35:12 UTC
252ab75 Frank Seide 15 December 2015, 19:35:53 UTC (comment) 15 December 2015, 19:35:53 UTC
dcb6c1d Frank Seide 15 December 2015, 19:35:37 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/mblayout 15 December 2015, 19:35:37 UTC
c54938f Dong Yu 15 December 2015, 18:34:10 UTC replace outdated ICASSP tutorial with more recent NIPS tutorial. 15 December 2015, 18:34:10 UTC
f6ef87b Frank Seide 15 December 2015, 17:36:51 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/mblayout 15 December 2015, 17:36:51 UTC
07b058b Frank Seide 15 December 2015, 17:36:25 UTC bug fix in RowStackNode: should not validate column dimension in presence of an MBLayout; bug fix: ComputationNode::InferMBLayoutFromInputsForStandardCase() had a the layout-consistency check commented out 15 December 2015, 17:36:25 UTC
48af1ef pkranen 15 December 2015, 17:09:25 UTC Fixed build-and-test script for cygwin local execution 15 December 2015, 17:09:25 UTC
81c6a1e Frank Seide 15 December 2015, 16:17:16 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/mblayout 15 December 2015, 16:17:16 UTC
52e04b1 Gaizka Navarro 15 December 2015, 14:20:06 UTC Added missing control files to solution 15 December 2015, 14:20:06 UTC
dba5f36 Gaizka Navarro 15 December 2015, 13:58:40 UTC Applied CR Feedback 15 December 2015, 14:16:14 UTC
4df3ed2 Gaizka Navarro 15 December 2015, 08:35:32 UTC Removed unused fixture. Updated environment variable error message 15 December 2015, 14:11:47 UTC
449a84f Philipp Kranen 15 December 2015, 10:04:41 UTC Removed files from VS solution that are not in Git anymore 15 December 2015, 10:04:41 UTC
b7c40c3 Philipp Kranen 15 December 2015, 08:58:11 UTC Removed Mixed Platforms and Win32 15 December 2015, 08:58:11 UTC
d8ffea0 Frank Seide 15 December 2015, 02:05:35 UTC (comments on DataTensor) 15 December 2015, 02:05:35 UTC
d273078 Frank Seide 15 December 2015, 00:13:51 UTC merged with master 15 December 2015, 00:13:51 UTC
dc32e4f Frank Seide 15 December 2015, 00:10:07 UTC (comments) 15 December 2015, 00:10:07 UTC
7a57f8b Frank Seide 14 December 2015, 23:40:50 UTC (added logging to calgammaformb() to track down an error) 14 December 2015, 23:40:50 UTC
f4a9155 Frank Seide 14 December 2015, 22:48:26 UTC bug fix, MBLayout::IsEnd() 14 December 2015, 22:48:26 UTC
af41546 Yongqiang Wang 14 December 2015, 21:12:36 UTC Merge branch 'master' of https://git.codeplex.com/cntk Conflicts: MachineLearning/CNTKSGDLib/DataReaderHelpers.h 14 December 2015, 21:12:36 UTC
e9fb669 Frank Seide 14 December 2015, 21:11:51 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/mblayout 14 December 2015, 21:11:51 UTC
abbc28e Frank Seide 14 December 2015, 21:11:29 UTC Coup de grâce: MBLayout reimplemented: the old bitmap-based implementation is gone. MBLayout now also stores sequence ids (so we can line up multiple MBLayouts, instrumental to sequence-to-sequence models), and it naturally allows for testing whether a delayed value cross a sequence boundary; deleted definition of MinibatchPackingFlags enum; old Is() and Get() functions are fully replaced by IsGap() and IsBeyondStartOrEnd(); HasGaps() test is now a one-liner 14 December 2015, 21:11:29 UTC
5ba34cf Frank Seide 14 December 2015, 19:43:48 UTC also removed generic MBLayout::Get(). Now tests are always for Gap or sequence boundary; bug fix: ImageReader must now initialize MBLayout with InitAsFrameMode() 14 December 2015, 19:43:48 UTC
cf931e2 Frank Seide 14 December 2015, 19:35:49 UTC removed MBLayout::Is(...t...) and Get(..t...). Now one must use Is(FrameRange), or some places can be more efficient now by avoiding to use this at all; new methods MBLayout::HasSequenceBeyondBegin() and likewise for End, for testing whether state needs to be preserved across minibatches 14 December 2015, 19:35:49 UTC
fe3076f Alexey Kamenev 14 December 2015, 19:33:26 UTC Fixed ImageReader and updated sample. 14 December 2015, 19:33:26 UTC
f961994 Jasha Droppo 14 December 2015, 17:52:15 UTC Merge branch 'master' into CUDA-elementwise-rework 14 December 2015, 17:52:15 UTC
25c5a12 Frank Seide 14 December 2015, 17:01:57 UTC deleted MBLayout::Set() and the old flag matrices--yay; deleted DelayedValueNodeBase::m_shiftedMBLayout (in lieu of the new MBLayout implementation) 14 December 2015, 17:01:57 UTC
0d808a5 yzhang87 14 December 2015, 16:04:20 UTC Merge branch 'master' of https://git.codeplex.com/cntk 14 December 2015, 16:04:20 UTC
b8eb51d yzhang87 14 December 2015, 16:00:20 UTC Fix kaldi Reader. 1) Using the new MBLayout interface. 2) Modify the configParameter to configRecordType to consistent with HTKMLFReader. 3) Clean the warning message. Next: refactor to make it consistent with HTKMLFReader. 14 December 2015, 16:00:20 UTC
8559820 Eldar Akchurin 14 December 2015, 15:05:29 UTC Fixing boost version 14 December 2015, 15:05:29 UTC
e573dd7 Eldar Akchurin 14 December 2015, 14:07:34 UTC Fixing greater and less than symobls 14 December 2015, 14:07:34 UTC
7c251eb Eldar Akchurin 14 December 2015, 13:56:49 UTC Fixing special characters 14 December 2015, 13:56:49 UTC
0bfa4c8 Eldar Akchurin 14 December 2015, 12:07:56 UTC Changing notification message for boost installation 14 December 2015, 12:07:56 UTC
83c94b0 Jasha Droppo 14 December 2015, 03:44:22 UTC CUDA removing redundant code in GPUMatrix::InplaceTruncate*() 14 December 2015, 03:44:22 UTC
8b49953 Jasha Droppo 14 December 2015, 02:18:28 UTC CUDA more elementwise kernel cleanup 14 December 2015, 02:18:28 UTC
5b02a5c Jasha Droppo 14 December 2015, 00:50:59 UTC CUDA kernel proposed fix - eliminate bad _assignSigmoidOf 14 December 2015, 00:50:59 UTC
642947a Frank Seide 13 December 2015, 21:11:11 UTC Merge branch 'master' of https://git.codeplex.com/cntk into fseide/mblayout 13 December 2015, 21:11:11 UTC
e066467 Frank Seide 13 December 2015, 21:07:52 UTC ColumnRangeWithMBLayoutFor(), ValueFor() etc. now understand FrameRange::m_timeOffset. DelayedValueNode has been changed to use this mechanism 13 December 2015, 21:07:52 UTC
88fe6b2 Frank Seide 13 December 2015, 20:19:27 UTC DelayedValueNode no longer uses m_pShiftedMBLayout (still kept around for use inside asserts to test the new code) 13 December 2015, 20:19:27 UTC
e6f8108 Jasha Droppo 13 December 2015, 18:55:46 UTC CUDA bringing changes in elementwise functions to NoGPU.cpp 13 December 2015, 19:46:02 UTC
d0e587d Jasha Droppo 13 December 2015, 17:59:18 UTC CUDA fix bug I accidentallty introduced in _scaleAndAddScalar kernel 13 December 2015, 19:45:59 UTC
back to top