https://github.com/Microsoft/CNTK

sort by:
Revision Author Date Message Commit Date
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
9041489 Fixing non existing sequence 12 January 2018, 18:59:38 UTC
6fc83ca Ensure lattice sequences matche the label sequences 12 January 2018, 18:59:38 UTC
8ab9d94 Fix indexing of lattice sequences 12 January 2018, 18:59:38 UTC
815e3fc Fixing definesMbSize for HTK deserializer 12 January 2018, 18:59:37 UTC
cf28412 update preferred device for lattice 12 January 2018, 18:59:37 UTC
f5edf14 Some cleanup of the SE with lattice node 12 January 2018, 18:59:37 UTC
a01da0c Fix lattice index for last sequence 12 January 2018, 18:59:37 UTC
3dba770 Testing parallel sequences 12 January 2018, 18:59:37 UTC
8c8560b Initial support of multiple lattice sequences 12 January 2018, 18:59:37 UTC
c36448c Able to run for single sequence 12 January 2018, 18:59:36 UTC
71880a9 Able to run up to lattice consupmtion in the SE node 12 January 2018, 18:59:36 UTC
98594d5 Enabled more SE parameters 12 January 2018, 18:59:36 UTC
6e485ad Further work to enable lattice deserializer in Sequence Node 12 January 2018, 18:57:27 UTC
30bf710 Make the new lattice node inherit from the legacy one 12 January 2018, 18:57:27 UTC
113cdf0 Initial work on deserialization of the lattice 12 January 2018, 18:57:27 UTC
589aa30 Adding binary packing 12 January 2018, 18:57:27 UTC
8aa9342 Add symlistpath as a parameter to the SE node' 12 January 2018, 18:57:26 UTC
b209dd0 Refactor lattice methods 12 January 2018, 18:57:25 UTC
fe31b52 Ensure that lattice metadata is passed inside the node 12 January 2018, 18:57:25 UTC
b2edf0c Rename lattice file parameters 12 January 2018, 18:57:25 UTC
5d7e995 Add parameters for the lattice symlit 12 January 2018, 18:57:24 UTC
4a96f60 Ensure lattice consumption in MB 12 January 2018, 18:57:24 UTC
1b2a256 Add SequenceWithLatticeNode 12 January 2018, 18:57:23 UTC
5a01480 Update the lattice reader sample size 12 January 2018, 18:51:50 UTC
4731e40 Update SGD to not consume lattice for new serializer 12 January 2018, 18:51:50 UTC
b88cac1 Refine lattice reader changes 12 January 2018, 18:51:50 UTC
f9539f8 make the initial version compile 12 January 2018, 18:51:50 UTC
e43e23f Lattice reader, add lattice index 12 January 2018, 18:51:50 UTC
31c8c1e Lattice deserializer - index 12 January 2018, 18:51:50 UTC
e1f1101 Lattice reader, cont 12 January 2018, 18:51:50 UTC
2758a8c Lattice reader, cont 12 January 2018, 18:51:49 UTC
0b76ebc Introduce lattice reader 12 January 2018, 18:51:49 UTC
959f1c8 Fixes #2704 (#2805) 12 January 2018, 08:34:03 UTC
8b261b8 Integrate sptiwari/group_conv_layer_bugfix into master 10 January 2018, 03:07:53 UTC
7142995 Allow cntk.times operator over tensors: [shape, free_axis] x [free_axis, shape] and [shape, batch_axis] x [batch_axis, shape]. 09 January 2018, 01:26:12 UTC
a04c1c9 Adding error message for Convolution layer when groups > 1. 09 January 2018, 01:04:35 UTC
21c4d20 Update current_iteration.md 08 January 2018, 18:43:27 UTC
ae2678f Integrate nikosk/synced-xforms3 into master 06 January 2018, 07:05:22 UTC
13ab006 Integrate vadimma/StopGradientBS into master 05 January 2018, 13:14:14 UTC
9bfe525 Bump up model version to 29 05 January 2018, 05:09:51 UTC
6ee82cc transforms with composite source work correctly. closes #2309 closes #2332 closes #2636 03 January 2018, 23:29:01 UTC
8bb6864 Expose StopGradient node in BS 03 January 2018, 19:26:06 UTC
6477d5b gather over axis 03 January 2018, 18:07:17 UTC
5214e2e Integrate v-yuxgu/crosstalkcaffe-bugfix into master 03 January 2018, 09:54:05 UTC
14105a2 Update current_iteration.md 03 January 2018, 06:57:12 UTC
3889d0a Update current_iteration.md 02 January 2018, 21:55:38 UTC
0e069fa fix bugs in crosstalkcaffe 02 January 2018, 09:35:22 UTC
348773e add missing ops (logical, reduceL1, reduceL2, etc.) with ONNX support 31 December 2017, 19:47:50 UTC
e4f8054 Fix IndexOutOfBound error in make_binary_drops 29 December 2017, 00:25:07 UTC
94e6582 Fix max retries while downloading model file Fix max retries to download FlappingBird_model.h5.model file. 21 December 2017, 18:39:44 UTC
f86f071 Update ONNX Reshape op to support InferredDimension. 19 December 2017, 20:04:15 UTC
5f359c7 Integrate eldak/fixingMaskForEval into master 18 December 2017, 21:20:12 UTC
accac8a Fixing Mask for eval 18 December 2017, 12:29:34 UTC
7e1e591 Making prefetch of several chunks in async manner 18 December 2017, 09:08:59 UTC
1fdea21 Revert c6f2f2c3d7952d7325fe2fc048a9669f7b86b325 14 December 2017, 19:15:43 UTC
c7d2502 squeeze, expand_dims, zeros_like, ones_like closes #2306 13 December 2017, 17:58:51 UTC
3cee5a0 add LogSoftmax, HardSigmoid, Flatten, Mean ops and ONNX support 12 December 2017, 16:03:19 UTC
d216454 fix reshape bug in backprop 12 December 2017, 03:12:39 UTC
4bb4dbb Adding DepthToSpace and SpaceToDepth ops to CNTK and ONNX. 09 December 2017, 06:45:28 UTC
f4f86cb Integrate liqun/onnxBugFixStage into master 07 December 2017, 07:25:54 UTC
5c97bd0 add top k operation. Closes #2468. 07 December 2017, 02:19:02 UTC
17e7e79 fix batch axis with broadcast issues 07 December 2017, 01:17:23 UTC
3dc66d0 Integrate sptiwari/onnx_convtranspose into master 07 December 2017, 00:16:48 UTC
3df734b Updating ONNX ConvTranspose to match spec. 06 December 2017, 22:05:09 UTC
95d8873 Bump version to 2.3.1 and update release notes. commit b906b2e94748f0303f4e4f41e860d882ab434a57 Author: Manik Jindal <manikj@microsoft.com> Date: Tue Dec 5 18:35:02 2017 -0800 Add whl and NuGet package links in README.md commit 5f4ff237c5ee3a449577ec29cbf93d64273c562a Author: Manik Jindal <manikj@microsoft.com> Date: Tue Dec 5 09:48:37 2017 -0800 Bump version 2.3 to 2.3.1 commit 1d0f37072524ad1a6de36bedf43359e0596c7382 Author: Manik Jindal <manikj@microsoft.com> Date: Tue Dec 5 00:26:26 2017 -0800 Add release notes commit 78b941b3e4e8c345bd6e8e45b6180ff624756d27 Author: Manik Jindal <manikj@microsoft.com> Date: Tue Dec 5 00:19:23 2017 -0800 Bump version from 2.3 to 2.3.1 06 December 2017, 05:15:50 UTC
c6f2f2c Integrate ec2cd71 into master 06 December 2017, 00:50:43 UTC
36ee9ce Integrate kedeng/fixLoss into master 05 December 2017, 23:31:08 UTC
6e15d5a Integrate 4074c0d58c91ba23d188526cf4281123ab47a7c0 into master 05 December 2017, 20:08:56 UTC
fd74a5c Adding C header to the drop 05 December 2017, 08:46:22 UTC
3d24a1b By Spandan add Pad Op Undo some changes to make resnet3d and resnet20 pass LotusRT 05 December 2017, 03:05:58 UTC
a78fb2b Fix incorrect axis comparison in cosine_distance_with_negative_samples and weighted_binary_cross_entropy 04 December 2017, 23:54:43 UTC
6a3e5c0 Integrate sptiwari/onnx_matmul into master 04 December 2017, 09:33:18 UTC
fdd21ee fix missing ONNX Ops; handle batch axis; ONNX ops unit test 04 December 2017, 05:19:52 UTC
e716074 Bug fix to create ONNX InputVariable with only one batch axis. 03 December 2017, 22:31:16 UTC
ec2cd71 Modified library import for "distributed" 03 December 2017, 07:56:26 UTC
e196c00 Merge branch 'master' into kyiwasak/faster-rcnn-distributed 03 December 2017, 07:49:59 UTC
4a49513 Integrate eldak/exposingCWrapperForHalide into master 01 December 2017, 16:47:37 UTC
d1c8643 Integrate jaliyaek/quantization_fix into master 01 December 2017, 16:32:13 UTC
f0d63b2 Exposing EvaluatorWrapper interface and restructure the code to be able to reuse most of the stuff in Halide inference engine 01 December 2017, 10:26:21 UTC
d948ffd Adding producer_name and producer_version fields to ONNX models. 01 December 2017, 00:42:47 UTC
97300ef quantizaion fix for convolutions with no bias parameter 30 November 2017, 20:00:40 UTC
01eb0fb Integrate liqun/onnxirintStage2 into master 30 November 2017, 07:04:36 UTC
744871f integrate with LotusRT 30 November 2017, 01:51:27 UTC
1d3d1e8 Implements batch normalization forward and backward in MKL. CPU training with BN is now enabled on Intel CPUs. 30 November 2017, 01:12:10 UTC
77de951 Integrate kedeng/fixKeras into master 29 November 2017, 21:48:23 UTC
bd11fbd Integrate eldak/jdEvalC into master 29 November 2017, 17:54:11 UTC
07fcce9 First version of C API for evaluation, taking over Jacob's branch. 29 November 2017, 15:16:38 UTC
4b7b7a8 Fixing restore in place with wrongly reordered placeholders 29 November 2017, 10:34:44 UTC
bdfb517 Integrate sptiwari/onnx_transpose into master 29 November 2017, 04:33:36 UTC
42587a8 Fix MKL convolution for output with reduced ranks 29 November 2017, 03:00:06 UTC
4028097 Handling the case when neither auto_pad nor pads is specified in Conv or Pooling ONNX nodes. 29 November 2017, 02:04:04 UTC
154e5af Integrate xiaoyongzhu-master into master 29 November 2017, 01:26:34 UTC
ef904fd Updating ONNX Transpose implementation to match updated spec. 29 November 2017, 00:17:22 UTC
back to top