| f34b7d7 | Vadim Mazalov | 31 August 2016, 21:01:06 UTC | Prevent repetitive update of the same node, which happens for recursive models. | 31 August 2016, 21:01:06 UTC |
| 9d6345b | Frank Seide | 24 August 2016, 15:05:13 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into fseide/kdd | 24 August 2016, 15:05:13 UTC |
| e27af11 | Project Philly | 24 August 2016, 07:47:02 UTC | Integrate zhouwang/rebase-mtNonUniform2 into master | 24 August 2016, 07:47:02 UTC |
| 835985b | Frank Seide | 24 August 2016, 06:39:28 UTC | fixed test | 24 August 2016, 06:39:28 UTC |
| bd5f9cb | Frank Seide | 24 August 2016, 04:54:00 UTC | addressed CR feedback; modifed an e2e test to cover inferInputRankToMap | 24 August 2016, 04:54:00 UTC |
| 7aa9b83 | Frank Seide | 24 August 2016, 01:12:36 UTC | merged from master | 24 August 2016, 01:12:36 UTC |
| 3c5aa75 | Frank Seide | 24 August 2016, 00:59:31 UTC | changed inferInputRank to inferInputRankToMap, and updated {Linear,Dense}Layer{} to accept the mutually exclusive inputRank and mapRank parameters | 24 August 2016, 00:59:31 UTC |
| de864bf | Frank Seide | 24 August 2016, 00:32:03 UTC | renamed Times inputRank to inferInputRank | 24 August 2016, 00:32:03 UTC |
| 86ec025 | Zhou Wang | 17 August 2016, 13:27:15 UTC | Use mt19937 instead of rand() | 23 August 2016, 17:24:45 UTC |
| 705951a | Zhou Wang | 11 August 2016, 13:02:59 UTC | adjust baseline with new boost random distribution function change std random distribution to boost random distribution | 23 August 2016, 17:23:46 UTC |
| 3179559 | Zhou Wang | 11 August 2016, 12:40:58 UTC | use boost random distribution functions instead of std | 23 August 2016, 17:23:42 UTC |
| c11ad31 | Zhou Wang | 09 August 2016, 13:51:57 UTC | Using slash instead of backslash | 23 August 2016, 17:23:26 UTC |
| a2565ab | Zhou Wang | 08 August 2016, 19:05:50 UTC | fix bug that removes path prefix in the file name on Linux Use comparision with tolerance use '/' instead of '\' in path in data files. | 23 August 2016, 17:22:47 UTC |
| 6744b96 | Zhou Wang | 08 August 2016, 12:52:00 UTC | check output with tolerance optionally allow tolerance when comparing file content | 23 August 2016, 17:21:43 UTC |
| aeee8f6 | Momcilo Vasilijevic | 09 June 2016, 14:12:24 UTC | Add bilinear initialization of learnable parameter Enable filling learnable parameter with coefficients for bilinear interpolation. A common use is with deconvolution acting as upsampling. | 23 August 2016, 15:26:25 UTC |
| d4f40dd | jeanfad | 23 August 2016, 08:21:42 UTC | adapt the swig wrapper to the change in getNextMinibatch method | 23 August 2016, 08:21:42 UTC |
| 10271ba | jeanfad | 23 August 2016, 07:38:24 UTC | Merge branch 'master' into blis/PythonV2 | 23 August 2016, 07:38:24 UTC |
| f46e911 | jeanfad | 23 August 2016, 07:37:03 UTC | move nn.py to examples folder | 23 August 2016, 07:37:03 UTC |
| abe6624 | Frank Seide | 23 August 2016, 00:35:15 UTC | re-fixed baselines from ErrorPrediction to ClassificationError | 23 August 2016, 00:35:15 UTC |
| 320f2f5 | Frank Seide | 23 August 2016, 00:27:23 UTC | re-merged baseline files from master, seems I mixed them up | 23 August 2016, 00:27:23 UTC |
| f32db3e | Frank Seide | 23 August 2016, 00:11:23 UTC | undid the test case of BrainScript from ClassificationError to ErrorPrediction | 23 August 2016, 00:11:23 UTC |
| ed7e602 | Frank Seide | 22 August 2016, 21:53:27 UTC | reapplied renaming from ErrorPrediction to ClassificationError for baseline files after merge | 22 August 2016, 21:53:27 UTC |
| 5b969ba | Frank Seide | 22 August 2016, 21:36:28 UTC | merged from master. Undid the ClassificationError baseline updates due to merge conflicts | 22 August 2016, 21:36:28 UTC |
| 1e68b3c | Frank Seide | 22 August 2016, 20:32:19 UTC | changed some more baselines from ErrorPrediction to ClassificationError | 22 August 2016, 20:32:19 UTC |
| 363f1e5 | Frank Seide | 22 August 2016, 20:27:40 UTC | changed some more baselines from ErrorPrediction to ClassificationError | 22 August 2016, 20:27:40 UTC |
| 8493f11 | Amit Agarwal | 22 August 2016, 18:54:21 UTC | CNTK v2 library: Clone constants to right device when user supplied value is not on the target compute device | 22 August 2016, 18:54:21 UTC |
| ba01512 | Frank Seide | 22 August 2016, 18:23:09 UTC | updated MatrixQuantizerTests to new >1-bit SGD defaults; addressed CR feedback | 22 August 2016, 18:23:09 UTC |
| 536e264 | Frank Seide | 22 August 2016, 18:16:39 UTC | renamed ErrorPrediction to ClassificationError, second attempt | 22 August 2016, 18:16:39 UTC |
| 37b6897 | Amit Agarwal | 22 August 2016, 17:48:54 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/cntkv2Library | 22 August 2016, 17:48:54 UTC |
| 9bd5738 | Frank Seide | 22 August 2016, 17:30:17 UTC | renamed ErrorPrediction to ClassificationError | 22 August 2016, 17:30:17 UTC |
| 1fa5b23 | jeanfad | 22 August 2016, 16:41:27 UTC | resnet cfair first complete network run | 22 August 2016, 16:41:27 UTC |
| 9521d7f | jeanfad | 22 August 2016, 14:34:41 UTC | move layer creation functions to a separate module | 22 August 2016, 14:34:41 UTC |
| 4df1389 | jeanfad | 22 August 2016, 14:13:28 UTC | convert the loss value to float | 22 August 2016, 14:13:28 UTC |
| d2eed70 | jeanfad | 22 August 2016, 14:12:41 UTC | turn mnist and ffnet python example into py.test tests | 22 August 2016, 14:12:41 UTC |
| 4e174f2 | Project Philly | 22 August 2016, 13:45:48 UTC | Integrate eldak/mtNonUniform2 into master | 22 August 2016, 13:45:48 UTC |
| a1528b9 | jeanfad | 22 August 2016, 13:22:54 UTC | fix to_numpy in swig so it reversed the shape to row major, and fix the shape reversal in ndshape.dimensions() | 22 August 2016, 13:22:54 UTC |
| 6a063e2 | Project Philly | 22 August 2016, 13:01:46 UTC | Integrate zhouwang/fix-deviceid-threadlocal into master | 22 August 2016, 13:01:46 UTC |
| de1c1e5 | jeanfad | 22 August 2016, 12:23:48 UTC | adapt the python api to the new Learner c++ api | 22 August 2016, 12:23:48 UTC |
| 941931d | jeanfad | 22 August 2016, 09:29:14 UTC | Merge branch 'master' into blis/PythonV2 | 22 August 2016, 09:29:14 UTC |
| e34528e | jeanfad | 22 August 2016, 09:24:55 UTC | fix file paths in examples, in order to work with py.test | 22 August 2016, 09:24:55 UTC |
| 6566012 | Zhou Wang | 22 August 2016, 08:04:47 UTC | Add comment to use thread_local after uprade to VS2015 | 22 August 2016, 08:04:47 UTC |
| d5f47f1 | Eldar Akchurin | 10 August 2016, 16:09:01 UTC | Updating sources and baselines to use MT | 22 August 2016, 07:49:08 UTC |
| b856ab7 | Eldar Akchurin | 10 August 2016, 10:43:58 UTC | Adding support for mersenne twister rand | 21 August 2016, 19:08:25 UTC |
| 45f4cbb | Amit Agarwal | 21 August 2016, 11:25:53 UTC | CNTK v2 library: Linux build fixes | 21 August 2016, 11:25:53 UTC |
| 668b6a4 | Amit Agarwal | 21 August 2016, 10:52:37 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/cntkv2Library | 21 August 2016, 10:52:37 UTC |
| fa4b99d | Amit Agarwal | 21 August 2016, 10:49:03 UTC | CNTK v2 Library: a) Add dynamic axis support b) New primitive functions and some higher level functions and c) Sequence classification test | 21 August 2016, 10:49:03 UTC |
| 050a840 | Frank Seide | 21 August 2016, 08:33:01 UTC | tuned >1-bit SGD: odd #quantization levels, range now 4 stddevs (before: 5) | 21 August 2016, 08:33:01 UTC |
| 9b4d27e | Frank Seide | 20 August 2016, 19:09:45 UTC | changed seqcla.cntk back to "gaussian" init, to be able to keep the baseline file | 20 August 2016, 19:09:45 UTC |
| deefdf2 | Frank Seide | 20 August 2016, 15:30:22 UTC | implemented ReduceElements ("mean"); updated LayerNormalizationLayer{} accordingly, no longer needs the 'dim' parameter; disabled first 2 of 3 logs of configuration variables in Release builds | 20 August 2016, 15:30:22 UTC |
| 8f07b8b | Frank Seide | 20 August 2016, 14:05:17 UTC | moved DumpNodeInfo() to Actions.h, now called DoDumpNodes() | 20 August 2016, 14:05:17 UTC |
| d2391a9 | Frank Seide | 20 August 2016, 07:02:10 UTC | changed EmbeddingLayer{} and LSTMBlock{} to 'heNormal' initialization | 20 August 2016, 07:02:10 UTC |
| db74d6b | Frank Seide | 20 August 2016, 06:34:17 UTC | changed ImageHandsOn from "gaussian" to "heNormal" initialization, and also most layers defaults in CNTK.core.bs | 20 August 2016, 06:34:17 UTC |
| e343e71 | Frank Seide | 20 August 2016, 05:25:48 UTC | added initOutputRank to ConvolutionalLayer{}, but in currently commented-out branch (needs another fix) | 20 August 2016, 05:25:48 UTC |
| 0e34934 | Frank Seide | 20 August 2016, 04:58:13 UTC | moved DiagTimes() to DeprecatedNodes.h, since it is redundant (ElementTimes) and should no longer be used; removed a log message from SeqCla's required test patterns as they are no longer being generated | 20 August 2016, 04:58:13 UTC |
| 5d6785f | Frank Seide | 20 August 2016, 04:19:25 UTC | bug fix in Times()-dimension inference: new parameter 'inputRank' was needed to guide inference of tensor rank; ReduceSum() etc. now understand axis=None; new global BS symbol Inferred to denote an inferred axis dimension | 20 August 2016, 04:19:25 UTC |
| f24b448 | Frank Seide | 20 August 2016, 02:04:43 UTC | new option initOutputRank to LearnableParameter to allow to control how fanIn and fanOut for random init are determined | 20 August 2016, 02:04:43 UTC |
| a236139 | Cha Zhang | 19 August 2016, 19:22:31 UTC | Bug fix for format checking. | 19 August 2016, 23:16:11 UTC |
| d52a02c | Cha Zhang | 19 August 2016, 17:00:44 UTC | Fix bug on n-dimensional convolution input format checking. | 19 August 2016, 23:16:11 UTC |
| 19d24e8 | Vadim Mazalov | 18 August 2016, 00:47:59 UTC | Ensure that network is created prior to reading features. This optimizes two scenarios: (1) allows to fail faster, when network definition has errors (2) when startEpoch==maxEpoch | 19 August 2016, 16:22:48 UTC |
| 80f08ae | jeanfad | 19 August 2016, 15:13:26 UTC | retrieve the loss value from the gpu | 19 August 2016, 15:13:26 UTC |
| 6f5ad32 | jeanfad | 19 August 2016, 14:53:49 UTC | first successful GPU training for CIFAR from Python v2 | 19 August 2016, 14:53:49 UTC |
| e9f4b8f | jeanfad | 19 August 2016, 14:53:01 UTC | fix pooling types | 19 August 2016, 14:53:01 UTC |
| 34a8d10 | jeanfad | 19 August 2016, 14:52:39 UTC | fix naming convention for some variables | 19 August 2016, 14:52:39 UTC |
| 921471a | jeanfad | 19 August 2016, 14:11:18 UTC | scale mnist input | 19 August 2016, 14:11:18 UTC |
| 008f2f1 | Eldar Akchurin | 19 August 2016, 13:11:44 UTC | Adding verbosity | 19 August 2016, 13:11:44 UTC |
| feccfcc | jeanfad | 19 August 2016, 12:36:55 UTC | fix relu name | 19 August 2016, 12:36:55 UTC |
| b0b80d0 | jeanfad | 19 August 2016, 12:36:35 UTC | fix bug in reverseing dimensions in swig | 19 August 2016, 12:36:35 UTC |
| 2f80057 | jeanfad | 19 August 2016, 12:35:29 UTC | remove semi colons | 19 August 2016, 12:35:29 UTC |
| d4275b4 | Eldar Akchurin | 19 August 2016, 11:49:48 UTC | Fixing linux build | 19 August 2016, 11:49:48 UTC |
| f27b1ce | Eldar Akchurin | 19 August 2016, 11:37:32 UTC | Small refactoring | 19 August 2016, 11:37:32 UTC |
| 702aa2d | Eldar Akchurin | 19 August 2016, 11:34:04 UTC | Hiding unsued variable | 19 August 2016, 11:34:04 UTC |
| 9165b43 | Eldar Akchurin | 19 August 2016, 11:32:56 UTC | Restructuring how zip index is built | 19 August 2016, 11:32:56 UTC |
| 3d80725 | Zhou Wang | 19 August 2016, 09:16:33 UTC | Define THREAD_LOCAL and force currentDevice to be THREAD_LOCAL | 19 August 2016, 09:16:33 UTC |
| e7d3894 | jeanfad | 18 August 2016, 14:17:35 UTC | change setup.py to release | 18 August 2016, 14:17:35 UTC |
| bb044d5 | jeanfad | 18 August 2016, 14:07:30 UTC | add epoch size var | 18 August 2016, 14:07:30 UTC |
| a51eb2e | jeanfad | 18 August 2016, 14:06:31 UTC | remove the tester line from swig readme | 18 August 2016, 14:06:31 UTC |
| e2b6cba | jeanfad | 18 August 2016, 13:55:59 UTC | remove an irrelevant comment from swig readme file | 18 August 2016, 13:55:59 UTC |
| 9bd9308 | Alexey Reznichenko | 18 August 2016, 12:33:38 UTC | Improve V2 Learner API, add basic serialization. FSAdaGrad, RMSProp: pre-allocate smoothed gradients with expected number of columns. Improve configuring built-in learners: add learning rates and momentums schedules. Add NDArrayView as a DictionaryValue type. Add tests for serialization and basic learner functionality. | 18 August 2016, 13:06:54 UTC |
| 2c457e2 | jeanfad | 18 August 2016, 10:41:34 UTC | cfarconverter py3 support | 18 August 2016, 10:41:34 UTC |
| c81a1a1 | jeanfad | 18 August 2016, 10:22:53 UTC | cifar_convert_py3 | 18 August 2016, 10:22:53 UTC |
| 244fa47 | jeanfad | 18 August 2016, 09:42:54 UTC | add timecmd batch file | 18 August 2016, 09:42:54 UTC |
| 03ea797 | jeanfad | 18 August 2016, 09:42:19 UTC | remove label map file for mnist | 18 August 2016, 09:42:19 UTC |
| 97abf8c | jeanfad | 18 August 2016, 09:38:05 UTC | fix data file for mnist pythonv2 example | 18 August 2016, 09:38:05 UTC |
| fbdd50a | jeanfad | 18 August 2016, 09:19:17 UTC | better handling of stream info in the examples | 18 August 2016, 09:19:17 UTC |
| 8018a33 | jeanfad | 18 August 2016, 09:03:11 UTC | remove fetch mnist data file | 18 August 2016, 09:03:11 UTC |
| 90d4159 | jeanfad | 18 August 2016, 09:01:21 UTC | remove label map file from mnist example | 18 August 2016, 09:01:21 UTC |
| 832a685 | Frank Seide | 17 August 2016, 16:54:09 UTC | Merge branch 'fseide/kdd' of https://github.com/Microsoft/cntk into fseide/kdd | 17 August 2016, 16:54:09 UTC |
| 9415713 | Frank Seide | 17 August 2016, 16:47:16 UTC | added None as a top-level BS constant; addressed CR feedback | 17 August 2016, 16:47:16 UTC |
| 32f97fd | Project Philly | 17 August 2016, 12:21:05 UTC | Integrate alrezni/output_fix into master | 17 August 2016, 12:21:05 UTC |
| 03a2be0 | Alexey Reznichenko | 17 August 2016, 09:43:57 UTC | A few minor changes in the BlockRandomizer output | 17 August 2016, 10:54:43 UTC |
| fa39bab | Wolfgang Manousek | 17 August 2016, 06:55:27 UTC | remove acml references from Kaldi readme | 17 August 2016, 08:04:56 UTC |
| 786ec99 | Wolfgang Manousek | 16 August 2016, 13:33:50 UTC | fixed broken ifdef statement | 17 August 2016, 08:04:56 UTC |
| 79cfcf7 | Wolfgang Manousek | 16 August 2016, 13:08:08 UTC | more acml removal | 17 August 2016, 08:04:56 UTC |
| 40d2478 | Wolfgang Manousek | 16 August 2016, 12:59:15 UTC | remove reference to acml math library | 17 August 2016, 08:04:56 UTC |
| 41fb7ad | Ivan Rodriguez | 09 August 2016, 09:31:47 UTC | Fix for linux build | 17 August 2016, 06:12:03 UTC |
| e69c28b | Ivan Rodriguez | 09 August 2016, 09:31:47 UTC | Delete actual dumps and fix dependency on Cudnn64_4.dll | 17 August 2016, 06:12:03 UTC |
| ba8f952 | Ivan Rodriguez | 08 August 2016, 14:59:05 UTC | Add test cases | 17 August 2016, 06:12:03 UTC |
| 3b6f687 | Ivan Rodriguez | 08 August 2016, 07:00:55 UTC | Modify brainscript tests projects to start with computation network tests | 17 August 2016, 06:12:03 UTC |
| d965072 | Zhou Wang | 16 August 2016, 15:49:06 UTC | fix bug in m_featureElmentType initialization | 16 August 2016, 15:49:06 UTC |
| 7aa981a | jeanfad | 16 August 2016, 15:01:58 UTC | add todo for ComputeInputPerDimMeansAndInvStdDevs | 16 August 2016, 15:01:58 UTC |
