https://github.com/Microsoft/CNTK

sort by:
Revision Author Date Message Commit Date
2436b8d Fix Linux only build breaks 15 February 2019, 22:47:33 UTC
939e799 Beautify code 15 February 2019, 21:58:01 UTC
2b26495 SGD Configure useFp16AllReduce 13 February 2019, 23:36:20 UTC
b354daa FP16 - Fix Runtime crash 05 February 2019, 17:11:51 UTC
37553c4 FP16 - Fix BMUF bug, update local copy of float aggregated weights for half 05 February 2019, 05:18:28 UTC
36095ee FP16 - Fix Linux build breaks 8 01 February 2019, 23:41:23 UTC
a74987f FP16 - Fix Linux build breaks 7 01 February 2019, 23:05:09 UTC
b8892fb FP16 - Fix Runtime error 2 01 February 2019, 21:39:44 UTC
d59a817 FP16 - Fix Runtime error 1 01 February 2019, 17:03:38 UTC
a4fc3f3 FP16 - Fix Linux build breaks 6 31 January 2019, 23:08:20 UTC
387ab37 FP16 - Brainscript smoothed gradients and checkpoint changes 31 January 2019, 21:51:58 UTC
518fd67 FP16 - Brainscript save load model 31 January 2019, 17:42:33 UTC
570803e FP16 - Fix Linux build breaks 5 31 January 2019, 16:51:45 UTC
1180125 FP16 - Fix Linux build breaks 4 30 January 2019, 22:55:20 UTC
b9bdfd1 FP16 - Fix build breaks 3 30 January 2019, 21:04:45 UTC
329b4be FP16 - Fix build breaks 2 30 January 2019, 20:52:57 UTC
340e704 FP16 - Fix build breaks 30 January 2019, 18:54:37 UTC
88471d4 FP16 - Fix Linux only build breaks 30 January 2019, 18:45:28 UTC
dfe4fa5 FP16 - Support fp16 in NDL network and SGD 29 January 2019, 21:24:12 UTC
615be0b Undo rightSplice change 18 December 2018, 20:57:31 UTC
3051731 Fix typo 2 18 December 2018, 01:21:38 UTC
01d7fa3 Fix typo 18 December 2018, 00:28:55 UTC
c1c55fc NVTX tracing with node name - 2 16 December 2018, 06:46:22 UTC
9cc3985 NVTX tracing with node name 16 December 2018, 01:22:27 UTC
fa93fdc Add NVTX tracer 15 December 2018, 01:48:07 UTC
51f8dbe Add dim label to bin mlf reader 15 December 2018, 00:37:58 UTC
43daca5 Expose binmlf in python 14 December 2018, 23:29:39 UTC
e26a3c1 Set rightSplice to 20 due to lack of plumbing to python 14 December 2018, 17:26:57 UTC
afc9c41 Reset local momentum to block momentum instead of 0.0 10 December 2018, 23:32:21 UTC
3b7761c Fix BMUF fp32 cache 04 December 2018, 20:23:31 UTC
06d9162 Fix to update fp32 cache after BMUF aggregate 04 December 2018, 16:11:48 UTC
b696b60 Restore bmuf fp16 codepath. No fp 32 cache. 30 November 2018, 15:03:07 UTC
bb44226 Fix BMUF bug - do NOT reset cached fp32 model in local momentum SGD leaner, ResetSmoothedGradients(). 30 November 2018, 08:36:56 UTC
1e30f9e Fix BMUF bug - save the very first model after local momentum SGD leaner casted to fp32. 30 November 2018, 03:43:27 UTC
bd6e9d0 More tracing 30 November 2018, 00:45:44 UTC
e2d52b4 Populate BMUF cached block model from Momentum SGD cached temp buffer 29 November 2018, 20:06:34 UTC
14f42f2 Change BMUF cached block model to fp32 28 November 2018, 00:50:25 UTC
b19d5e2 Fix crash 3 27 November 2018, 22:20:40 UTC
632c1b1 Fix crash 2 27 November 2018, 21:26:50 UTC
9e38cd4 Fix crash 27 November 2018, 18:46:39 UTC
b92b838 Use exact shape from matrix 27 November 2018, 00:58:43 UTC
bfe4a77 Fix BMUF fp16 bug. And disable fp16 allreduce for fp32. 27 November 2018, 00:46:41 UTC
532ae70 Init NCCL earlier 26 November 2018, 17:38:03 UTC
97d6c9a Remove extra logging 26 November 2018, 17:30:20 UTC
d2cfa9f fp16 aggregate - logging 21 November 2018, 22:08:49 UTC
c997405 fp16 aggregate - only on gpu - correct cast direction 21 November 2018, 18:38:49 UTC
3d30311 fp16 aggregate - only on gpu 21 November 2018, 18:22:08 UTC
e784f5b Add fp16 aggregation 20 November 2018, 18:16:05 UTC
ccb504d Adjust tracing 17 November 2018, 08:22:50 UTC
1657adb Merge pull request #3506 from Microsoft/BMUF-float16 Bmuf float16 16 November 2018, 19:17:34 UTC
275a14b Fix dumpnode crash 16 November 2018, 19:13:54 UTC
b00f140 Fix reserve to ptr not obj 16 November 2018, 19:12:16 UTC
569297b Adjust tracing timeout values 15 November 2018, 20:17:37 UTC
1c3a65d Add more tracing to HTK readers 2 15 November 2018, 06:58:38 UTC
ae9e41d Add more tracing to HTK readers 14 November 2018, 23:10:58 UTC
ccfb0a1 Add useV2Aggregator to SGD config 14 November 2018, 22:53:31 UTC
55294c4 UseV2Aggregator 13 November 2018, 19:05:50 UTC
54f8f14 Update DistributedCommunicator.cpp added log for fp16 aggregation 12 November 2018, 19:15:39 UTC
3e6eef5 Update BlockMomentumDistributedLearner.h corrected type for GetMatrix and GetWritableMatrix 09 November 2018, 18:02:55 UTC
17a02f4 Update BlockMomentumDistributedLearner.h 09 November 2018, 17:48:15 UTC
4f79190 Update BlockMomentumDistributedLearner.h updated BMUF --> ResetBuffer to accept float16 type 09 November 2018, 00:42:11 UTC
22c7c3c Update BlockMomentumDistributedLearner.h 08 November 2018, 22:32:13 UTC
69df29e Merge branch 'liqun/seqopsStage' 08 November 2018, 22:14:44 UTC
2405457 Update BlockMomentumDistributedLearner.h 08 November 2018, 20:27:55 UTC
ab4bee2 Support RNN ops in a Scan loop Update with latest ONNX Update with latest ONNX graph IR Support sequence ops - Sequence::Gather, Sequence::PastValue, Sequence::FutureValue, etc. 08 November 2018, 02:36:20 UTC
3f46cf0 Updates on several ONNX exports. * ConvTranspose outputShape: now pads values are always exported even when outputShape is given. The reason is that CNTK and ONNX have different padding specs. * Flatten: in CNTK flatten does not affect batch axis, this should be preserved in ONNX. 03 November 2018, 00:18:07 UTC
a19ce9e Resolve forward_backward docstring Sphinx error. 02 November 2018, 17:43:24 UTC
29818ff Merge branch 'bowbao/onnxruntime_ci_stage2' 01 November 2018, 06:58:24 UTC
0cd2fae Merge branch 'pull/3476' 31 October 2018, 23:43:42 UTC
fca1396 Add onnx_test_runner verification in CI. * onnx_test_runner.exe will be called on win64 GPU tests to verify if the output data produced by CNTK in onnx_op_test and onnx_model_test(cntk_model_test) matches in onnxruntime. 31 October 2018, 22:46:56 UTC
a0d6f14 Fixed type and error in identifying the deserialiser 31 October 2018, 13:05:27 UTC
cdd18d2 Update LSTM/Truncated-Kaldi CPU baseline 29 October 2018, 18:16:42 UTC
3cef1fd fix python 27 onnx test 28 October 2018, 17:22:07 UTC
1ff5548 Set linux pip install future version to 0.16.0 * This was not set explicitly. Lately the version was updated to 0.17.0 and is causing failure in CI. 26 October 2018, 17:40:01 UTC
0b41602 Update more baseline affected by fixing InvStdDev. 26 October 2018, 00:43:52 UTC
77dad35 Added better documentation to forward_backward 25 October 2018, 11:17:25 UTC
e940605 Support ONNX Scan op 20 October 2018, 04:36:21 UTC
a55e871 Fix InvStdDev. * Issue was that AssignSqrOfDifferenceOf(beta, input, mean, alpha) assigns mean value to the gaps in input. These values are then reduced within this function, leading to incorrect results. The fix is to execute assign and reduce separately, and mask gaps to zero again before reducing. * Update test baseline affected by this change (err is lowered by <1%). 19 October 2018, 17:24:11 UTC
0ffdcf7 Overhaul node name export & other fixes * Overhaul node name export. Create static class UniqueNodeNameStorage to manage ONNX node name generation with maintained one-to-one mapping between CNTK Uid, while preserving the original CNTK node name by best efforts (#3358). * Update onnx_op_test to test the preservation of original CNTK node names in exported/imported models. * Update onnx_test_helper to support proper linking of test data and onnx model input/output with unique names. * Update onnx_test_helper to generate .bat file to run exported models in further onnxruntime verification. * Fix Sum import to support arbitrary number of inputs. Sum implementation in CNTK backend is loop of Plus, which takes care of potential broadcast issues. 18 October 2018, 01:36:48 UTC
da2e610 Replaced wchar/wstring with char/string in C interface 12 October 2018, 21:12:23 UTC
149d87b Adding ONNX export support for OneHotOp. 05 October 2018, 21:06:56 UTC
6063866 Merge branch 'bowbao/pooling_pad_offset' 02 October 2018, 03:41:06 UTC
bf37aad Fix pad offset computation for pooling * Compute keyInterior according to the updated algorithm for computing cell offset key. * Update unittest of avg_pooling/max_pooling for cases that requires auto_padding = True. Previous test cases cover only those that do not need padding. 02 October 2018, 00:21:30 UTC
7c838d3 fix csharp examples and references 01 October 2018, 22:45:13 UTC
fcdeef6 Support crop_manual export & import. 29 September 2018, 20:46:17 UTC
a36fae8 Support logPlus(log_add_exp) export to ONNX * ONNX supports similar op ReduceLogSumExp. Conversions are added when exporting. * Refactored CNTKToONNXHelper::BroadcastInputsIfNeeded to support more generalized cases. 28 September 2018, 22:59:55 UTC
c2072cc Add support for ONNX export of StraightThrough op. 27 September 2018, 17:25:11 UTC
1aab76a Updating ONNX submodule hash to include defs for ConstantLike and EyeLike ops. 27 September 2018, 01:01:34 UTC
ce503f8 pooling export fix for backward compatibility 26 September 2018, 00:09:39 UTC
9165fd0 Merge branch 'kedeng/fixCrash' 25 September 2018, 00:26:34 UTC
58f810f update with ONNX1.3 and latest onnxruntime 22 September 2018, 16:53:27 UTC
1489de8 Fix a crash in transpose_times simplification to element times 22 September 2018, 05:33:41 UTC
93cc680 Update reshapeing_test to accomodate python 2.7 21 September 2018, 17:24:57 UTC
d626c2a Merge branch 'bowbao/gather_backward' 21 September 2018, 05:02:57 UTC
da6b0bc GatherNode backward: add check for no dynamic axis Previously, to resolve issue of gather producing incorrect gradient values, validity mask check was added to ensure we don't count non-valid cells as 0. However, this check is needed only for input that has dynamic axis, i.e. inputs that have MBLayout. 20 September 2018, 21:54:39 UTC
b2c28cc updating iteration documentation file 20 September 2018, 21:39:58 UTC
0a3eb3b Update onnx_model_test with tests on cntk pretrained models 19 September 2018, 18:27:55 UTC
6f09c39 Merge branch 'release/2.6' 18 September 2018, 02:33:30 UTC
4ed1896 set public_build to "no"/false 17 September 2018, 23:10:48 UTC
1be3b64 update readme for .net support 15 September 2018, 01:07:45 UTC
back to top