7bda219 | Amit | 15 February 2016, 21:03:31 UTC | Add arch=compute_30 codegen for debug builds required by CNTK native BatchNorm implementation | 15 February 2016, 21:03:31 UTC |
11f11b1 | vmazalov | 15 February 2016, 01:23:37 UTC | Add printMetadata config to documentation for the dumpNode top-level command. | 15 February 2016, 01:23:37 UTC |
f89e544 | Project Philly | 14 February 2016, 17:31:17 UTC | Integrate AddPrintMetadataFlag into master | 14 February 2016, 17:31:17 UTC |
e1093b9 | Project Philly | 13 February 2016, 15:28:20 UTC | Integrate f0d8a23b26cbfe94640aa9869dcf57e4a4352455 into master | 13 February 2016, 15:28:20 UTC |
d2b84a5 | Project Philly | 13 February 2016, 13:54:40 UTC | Integrate mahilleb/WindowsCpuOnlyTest into master | 13 February 2016, 13:54:40 UTC |
6ff885a | Alexey Kamenev | 12 February 2016, 19:43:40 UTC | Fixed build warning. | 12 February 2016, 19:43:40 UTC |
19b0c82 | Alexey Kamenev | 12 February 2016, 18:13:45 UTC | Addressed code review feedback. | 12 February 2016, 19:34:03 UTC |
9d6f68f | Mark Hillebrand | 12 February 2016, 19:33:14 UTC | Enable Windows CPU-only end-to-end tests (empty commit) | 12 February 2016, 19:33:14 UTC |
fe997db | Alexey Kamenev | 11 February 2016, 23:02:38 UTC | Fix for printf warning. | 12 February 2016, 17:34:31 UTC |
03e8cc5 | Alexey Kamenev | 11 February 2016, 22:20:34 UTC | Addressed code review feedback. | 12 February 2016, 17:34:17 UTC |
e12427d | Alexey Kamenev | 11 February 2016, 02:00:01 UTC | Fix for name resolution. | 12 February 2016, 17:34:02 UTC |
e8ab337 | Alexey Kamenev | 11 February 2016, 01:50:53 UTC | Refactored Makefile to include defines both to NVCC and g++. | 12 February 2016, 17:33:47 UTC |
e25e548 | Alexey Kamenev | 10 February 2016, 20:07:32 UTC | Added SM30 config to MathCuda Debug build defaults. | 12 February 2016, 17:33:32 UTC |
958f634 | Alexey Kamenev | 10 February 2016, 19:43:12 UTC | Fixed builds. | 12 February 2016, 17:33:18 UTC |
820cfc2 | Alexey Kamenev | 10 February 2016, 19:03:41 UTC | Fixed bug with BN engine param name in BS and fixed CPU build. | 12 February 2016, 17:33:03 UTC |
11fe2c6 | Alexey Kamenev | 10 February 2016, 01:02:43 UTC | Added eps and engine parameters to BN node. | 12 February 2016, 17:32:49 UTC |
9ad6147 | Alexey Kamenev | 09 February 2016, 21:11:56 UTC | Batch norm: added epsilon as parameter, enabled run mean/invstddev. | 12 February 2016, 17:32:35 UTC |
1379f58 | Alexey Kamenev | 09 February 2016, 07:35:18 UTC | Added convolutional implementation for batch norm. | 12 February 2016, 17:32:21 UTC |
5846fae | Alexey Kamenev | 06 February 2016, 02:07:56 UTC | Improved BN unit tests. | 12 February 2016, 17:32:07 UTC |
21575f3 | Alexey Kamenev | 05 February 2016, 06:29:12 UTC | Added batch norm per-activation backprop implementation. | 12 February 2016, 17:31:53 UTC |
4e9483b | Alexey Kamenev | 04 February 2016, 02:39:40 UTC | Added batch norm per-activation feed forward implementation. | 12 February 2016, 17:31:39 UTC |
6465c80 | Project Philly | 12 February 2016, 16:14:42 UTC | Integrate mahilleb/02_Convolution_Unstable into master | 12 February 2016, 16:14:42 UTC |
4e8c3e7 | Mark Hillebrand | 12 February 2016, 14:55:54 UTC | Tests/EndToEndTests/Examples/Image/MNIST/02_Convolution: unstable on Windows | 12 February 2016, 15:44:58 UTC |
7d24810 | Project Philly | 12 February 2016, 15:01:12 UTC | Integrate mahilleb/TestDriverWinCpuonly into master | 12 February 2016, 15:01:12 UTC |
f0d8a23 | Ross Wightman | 29 January 2016, 23:44:37 UTC | Add experimental support for OpenBLAS library * Add 'openblas' as mathlib option in configure. Not added to auto-search so must be specified using --with-openblas * configure script searches empty tail so that libraries located at default_path_list roots (ie /usr/local/ + include/openblas_config.h) are found * Treat ACML as the odd library out in ifdefs since it doesn't conform to typical BLAS standard. Other libraries like ATLAS should be able to share OpenBLAS/MKL variants. Add default USE_ACML define in VS projects to match * Fix 'max' macro define colliding with C++ std::max once openblas headers are included Usage Notes: * For best performance, build OpenBLAS with USE_OPENMP=1. When running CNTK, set OPENBLAS_NUM_THREADS environment var or set numCPUThreads CNTK config variable to the physical core count or performance will suffer * OpenBLAS 2.16 (git HEAD) tested in Linux with GCC 4.8.4 and in Windows with OpenBLAS 2.15 (pre-built binary release + MingGW 64-bit support dlls) * For Windows, in Math.vcxproj, replace libacml_mp_dll.lib with libopenblas.dll.a and change USE_ACML define to USE_OPENBLAS. Change ACML_PATH environment variable to your OpenBLAS path. Modify openblas_config.h as per https://github.com/xianyi/OpenBLAS/issues/708 * On current generation Intel processors, OpenBLAS measures a little faster than AMD ACML and slower than Intel MKL on MNIST and other examples | 11 February 2016, 21:40:18 UTC |
9386b8d | Amit Agarwal | 11 February 2016, 18:50:57 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/htkmlfreaderMemoryOptimizations | 11 February 2016, 18:50:57 UTC |
e511c9d | Amit Agarwal | 11 February 2016, 18:50:49 UTC | Addressed CR feedback | 11 February 2016, 18:50:49 UTC |
7652ef0 | Mark Hillebrand | 11 February 2016, 15:17:21 UTC | Merge pull request #90 from felixonmars/patch-1 Fix a typo: lanugage -> language | 11 February 2016, 15:17:21 UTC |
c360717 | Project Philly | 11 February 2016, 13:27:19 UTC | Integrate mahilleb/Msbuild2 into master | 11 February 2016, 13:27:19 UTC |
416577f | Eldar Akchurin | 11 February 2016, 09:49:08 UTC | Fixing the AV for the partial minibatch | 11 February 2016, 09:54:25 UTC |
c4ba87d | Project Philly | 11 February 2016, 08:41:16 UTC | Integrate cd4808e into master | 11 February 2016, 08:41:16 UTC |
58c768b | Amit Agarwal | 10 February 2016, 21:52:27 UTC | Temporarily retained old memory inefficient frame randomization for side by side testing. The old randomization is off by default and can be switched on using a config setting minimizeReaderMemoryFootprint=false | 10 February 2016, 23:26:58 UTC |
3021ad0 | Amit Agarwal | 10 February 2016, 19:53:23 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/htkmlfreaderMemoryOptimizations | 10 February 2016, 19:53:23 UTC |
4a3b6ea | Mark Hillebrand | 10 February 2016, 16:52:49 UTC | Tests/TestDriver.py: pick up Windows CPU-only binaries from x64/{Debug,Release}_CpuOnly | 10 February 2016, 16:52:49 UTC |
73b27ef | Mark Hillebrand | 10 February 2016, 11:18:24 UTC | Address CR comments | 10 February 2016, 11:18:24 UTC |
1b29d2f | Mark Hillebrand | 05 February 2016, 08:12:58 UTC | .vcxproj: let intermediate output go into a common directory Fixes #12. | 10 February 2016, 08:10:07 UTC |
2c2a721 | Mark Hillebrand | 04 February 2016, 21:18:18 UTC | CNTK.Cpp.props: define properties to distinguish build flavors and use them | 10 February 2016, 08:05:31 UTC |
ea55623 | Mark Hillebrand | 04 February 2016, 20:39:37 UTC | Introduce CNTK.Cpp.props and load in every .vcxproj | 10 February 2016, 08:05:31 UTC |
52aad32 | Mark Hillebrand | 04 February 2016, 20:17:47 UTC | .vcxproj: drop include path $(VCInstallDir)atlmfc\include | 10 February 2016, 08:05:31 UTC |
9c80220 | Mark Hillebrand | 04 February 2016, 20:09:00 UTC | .vcxproj: remove HTKMLFReader IncludePath where not needed | 10 February 2016, 08:05:31 UTC |
3b979f8 | Mark Hillebrand | 04 February 2016, 19:36:30 UTC | .vcxproj: coalesce importing for .user.props for Debug / Release | 10 February 2016, 08:05:31 UTC |
c2ccd66 | Mark Hillebrand | 04 February 2016, 19:29:09 UTC | .vcxproj: remove empty <Command/> | 10 February 2016, 08:05:31 UTC |
2b22008 | Mark Hillebrand | 04 February 2016, 19:15:13 UTC | .vcxproj: remove empty custom build steps | 10 February 2016, 08:05:31 UTC |
8ffd7e7 | Mark Hillebrand | 04 February 2016, 19:13:16 UTC | CNTK.vcxproj: fix: add missing parameter for prebuild.bat for Release builds | 10 February 2016, 08:05:31 UTC |
7baf28a | Amit | 10 February 2016, 06:18:35 UTC | Fixed linux build | 10 February 2016, 06:18:35 UTC |
2ec9d7a | Amit Agarwal | 10 February 2016, 05:29:54 UTC | Replace incorrect use of bad_exception with runtime_error exceptions | 10 February 2016, 05:29:54 UTC |
409327a | Amit Agarwal | 09 February 2016, 21:09:18 UTC | Write GPUMatrix to file in chunked manner to avoid creating very large intermediate CPU buffers | 09 February 2016, 21:09:18 UTC |
439707a | Amit Agarwal | 09 February 2016, 18:55:09 UTC | Added new handler to print call stack on allocation failure | 09 February 2016, 18:55:09 UTC |
57fd382 | Amit Agarwal | 09 February 2016, 05:43:09 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/htkmlfreaderMemoryOptimizations | 09 February 2016, 05:43:09 UTC |
5945c39 | Amit Agarwal | 09 February 2016, 00:42:23 UTC | Removed fsync call from fflushOrDie since its is causing issues when writing philly models to HDFS | 09 February 2016, 01:34:45 UTC |
e9f5ca4 | Amit Agarwal | 09 February 2016, 00:12:05 UTC | Write checkpoint file before writing model file, so that a checkpoint file is guaranteed to exist for the latest successfully written model file | 09 February 2016, 00:12:05 UTC |
5a67cf2 | Amit Agarwal | 08 February 2016, 21:10:49 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/htkmlfreaderMemoryOptimizations | 08 February 2016, 21:10:49 UTC |
c214980 | Amit Agarwal | 08 February 2016, 21:10:40 UTC | Disabled HTKMLFReader unit tests HTKMLFReaderSimpleDataLoop7 and HTKMLFReaderSimpleDataLoop17 as they use invalid config setting combination of Truncated=true and frameMode=true | 08 February 2016, 21:10:40 UTC |
a2b99a4 | Amit Agarwal | 08 February 2016, 00:49:40 UTC | HTKMLFReader memory optimization: Reimplemented frame mode randomization to keep just a window of randomized frames in memory instead of having a full array of all frames in the corpus | 08 February 2016, 00:49:40 UTC |
cd4808e | David Rolland | 07 February 2016, 05:46:35 UTC | Added missing quote in case OPENCV_PATH contains spaces. | 07 February 2016, 05:46:35 UTC |
bb608f8 | Frank Seide | 07 February 2016, 04:31:48 UTC | LSTM test now uses an actual 1D vector for the bias parameter (as opposed to a 1-column 2D matrix) | 07 February 2016, 04:31:48 UTC |
6887465 | Frank Seide | 07 February 2016, 04:05:57 UTC | fixed a size_t wraparound in TensorSliceWithMBLayoutFor() that broke some 1-column tensors; also sorted m_allRoots by name for better reproducability | 07 February 2016, 04:05:57 UTC |
f321625 | Amit Agarwal | 06 February 2016, 07:06:15 UTC | Added a check for inconsistent frameMode and Truncated settings. Also addresed some minor CR feedback. | 06 February 2016, 07:06:15 UTC |
9279662 | Frank Seide | 06 February 2016, 01:10:53 UTC | Matrix::Print() can now print corners for easier debugging | 06 February 2016, 01:10:53 UTC |
2f9a48c | Frank Seide | 05 February 2016, 19:06:20 UTC | InputNodes() now skips inputs that are only reachable through PreComputeNodes that have already been computed, addressing Issue #65; cleaned up some unnecessary NULL checks before delete | 05 February 2016, 19:06:20 UTC |
2bbee29 | U-FAREAST\fseide | 05 February 2016, 17:18:59 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into fseide/fixInputNodes | 05 February 2016, 17:18:59 UTC |
7c86176 | Frank Seide | 05 February 2016, 17:18:45 UTC | temporary fix for "computeGradient" parameter inconsistency | 05 February 2016, 17:18:45 UTC |
84e9db1 | Amit Agarwal | 05 February 2016, 02:29:31 UTC | Added a workaround for the HDFS FUSE implementation bug of failure to delete destination file when the destination file specified to rename function already exists | 05 February 2016, 02:29:31 UTC |
83e7783 | Frank Seide | 04 February 2016, 23:42:08 UTC | better error messages for Win32 exceptions | 04 February 2016, 23:42:08 UTC |
d3e636c | Amit Agarwal | 04 February 2016, 23:08:39 UTC | Fixed redundant writing of the .0 model file when no PreCompute is performed and the model is loaded from a checkpoint | 04 February 2016, 23:23:40 UTC |
9cf21b7 | U-FAREAST\fseide | 04 February 2016, 22:52:02 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into fseide/fixInputNodes | 04 February 2016, 22:52:02 UTC |
98a723c | Frank Seide | 04 February 2016, 22:51:53 UTC | reduced clutter of validation output by validating the entire network, rather than each output node individually | 04 February 2016, 22:51:53 UTC |
a930d1c | thhoens | 03 February 2016, 00:01:01 UTC | Updated EvalActions to fix support for OutputNodeNames in the config. | 04 February 2016, 21:11:13 UTC |
14881a2 | Amit Agarwal | 04 February 2016, 18:55:28 UTC | Added guards to all model saving calls to prevent mutiple parallel workers from trying to write the model file concurrently | 04 February 2016, 18:55:28 UTC |
1e65a41 | Frank Seide | 04 February 2016, 16:51:08 UTC | fixed bug in catching when MPI runtime is missing | 04 February 2016, 16:51:08 UTC |
17ea344 | Felix Yan | 04 February 2016, 14:20:50 UTC | Fix a typo: lanugage -> language | 04 February 2016, 14:20:50 UTC |
8b2c9ff | Project Philly | 04 February 2016, 12:22:24 UTC | Integrate mahilleb/ImageReaderOptDepFix into master | 04 February 2016, 12:22:24 UTC |
29dfe2d | Project Philly | 04 February 2016, 11:28:46 UTC | Integrate eldak/addressingReaderCR into master | 04 February 2016, 11:28:46 UTC |
7914c7c | Project Philly | 04 February 2016, 11:13:07 UTC | Integrate pkranen/fixMNIST into master | 04 February 2016, 11:13:07 UTC |
5fa84b0 | Mark Hillebrand | 03 February 2016, 23:16:58 UTC | ImageReader.vcxproj: don't build if OpenCV not present | 04 February 2016, 10:00:26 UTC |
f5a349a | Eldar Akchurin | 03 February 2016, 15:30:58 UTC | Addressing ImageReader CR from Frank | 04 February 2016, 08:37:14 UTC |
7d1fc36 | Philipp Kranen | 03 February 2016, 13:21:58 UTC | fixed .config to .cntk in CSEvalCient | 04 February 2016, 08:11:06 UTC |
576028b | Philipp Kranen | 03 February 2016, 12:56:58 UTC | fixed 01.. config to match wiki page | 04 February 2016, 08:11:04 UTC |
27fd4e0 | Amit Agarwal | 04 February 2016, 07:05:16 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/htkmlfreaderMemoryOptimizations | 04 February 2016, 07:05:16 UTC |
eb231c3 | Frank Seide | 04 February 2016, 00:52:47 UTC | (added a comment) | 04 February 2016, 00:52:47 UTC |
0900bdb | Frank Seide | 03 February 2016, 23:27:00 UTC | gave up back-compat with some mis-spelled names of training criterion nodes | 03 February 2016, 23:27:00 UTC |
78587f5 | Amit Agarwal | 03 February 2016, 20:19:54 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/htkmlfreaderMemoryOptimizations | 03 February 2016, 20:19:54 UTC |
070006d | Frank Seide | 03 February 2016, 19:49:13 UTC | addressed an issue discovered in static code analysis | 03 February 2016, 19:49:13 UTC |
0626c5d | U-FAREAST\fseide | 03 February 2016, 19:36:39 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into fseide/fixCrossProcessMutex | 03 February 2016, 19:36:39 UTC |
2d11980 | Frank Seide | 03 February 2016, 19:32:13 UTC | addressed bugs discovered by static code analysis, Issue #81, and manually merged two related pull requests | 03 February 2016, 19:32:13 UTC |
498e022 | Alexey Kamenev | 03 February 2016, 18:01:20 UTC | Fixed MNIST batch norm sample. | 03 February 2016, 18:01:50 UTC |
9f9b681 | Alexey Kamenev | 01 February 2016, 19:31:02 UTC | Added README.md with disclaimer to ResNet samples. | 03 February 2016, 18:01:49 UTC |
c394833 | U-FAREAST\fseide | 03 February 2016, 17:32:47 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into fseide/fixCrossProcessMutex | 03 February 2016, 17:32:47 UTC |
9b1f342 | Project Philly | 03 February 2016, 17:30:22 UTC | Integrate mahilleb/WindowsVsDependenciesDoc into master | 03 February 2016, 17:30:22 UTC |
87acb14 | Frank Seide | 03 February 2016, 17:24:09 UTC | final simplification of case-insensitive string comparisons | 03 February 2016, 17:24:09 UTC |
901f99f | Frank Seide | 03 February 2016, 17:17:58 UTC | further simplified case-insensitive string compares | 03 February 2016, 17:17:58 UTC |
16175a1 | Frank Seide | 03 February 2016, 17:01:55 UTC | cleanedup all case-insensitive comparisons | 03 February 2016, 17:01:55 UTC |
f454962 | Frank Seide | 03 February 2016, 16:38:24 UTC | fixed process-name comparison in BestGpu::QueryNvmlData() | 03 February 2016, 16:38:24 UTC |
7ab3d9e | Mark Hillebrand | 25 January 2016, 13:22:33 UTC | .vcxproj: Let diagnostics about dependencies point to CNTK Wiki | 03 February 2016, 16:22:13 UTC |
ee97506 | U-FAREAST\fseide | 03 February 2016, 16:09:37 UTC | Merge branch 'master' of https://github.com/Microsoft/CNTK into fseide/fixCrossProcessMutex | 03 February 2016, 16:09:37 UTC |
ab7d5f5 | Project Philly | 03 February 2016, 15:02:45 UTC | Integrate mahilleb/BuildAndTestTuneNew into master | 03 February 2016, 15:02:45 UTC |
fe75bb1 | Mark Hillebrand | 02 February 2016, 07:31:52 UTC | Extend build information included / reported by CNTK * Build target: GPU or CPU-only * Build type (Debug or Release), was missing for Windows * Whether or not 1bit-SGD was included * If there were local modifications in the Git working copy * Generate BUILD{ER,MACHINE,PATH} on Linux | 03 February 2016, 14:48:39 UTC |
0e7c663 | Mark Hillebrand | 16 December 2015, 23:52:34 UTC | Tools/build-and-test: fixes and extensions * Add --test-targets options to limit test to CPU / GPU CPU builds support testing only for CPU, GPU builds support testing for both. * Add --clean-build-after, hook-up and fix existing code. * Change build directories on Linux to build/$TARGET/$FLAVOR * Fix clean-up bug between test run. * More quoting for 'rm' * Support CPU-only build for Windows * MSbuild in parallel and with less output * Exit if both GPU and 1bit-SGD is requested on Windows (side-by-side not yet support) * GPU tests will specify device ID 0, not Auto | 03 February 2016, 14:39:14 UTC |
0e5ad89 | Clemens Marschner | 02 February 2016, 14:45:50 UTC | Reapply fix in Makefile for SparsePCReader Reapply name changes in Makefile that had been resolved but got lost during excessive rebasing. | 03 February 2016, 09:15:34 UTC |
d23eecc | Mark Hillebrand | 03 February 2016, 08:13:20 UTC | Fixed TestDriver.py output format regressions introduced in dc1ed4628f | 03 February 2016, 08:28:17 UTC |