4003c087a1
Merge fp16 brainscript work ( #3606 )
...
* FP16 BrainScript - address code review comments
* Remove Tab and fix debug build breaks
* Fix Linux Build breaks
* fp16 brain script - add _CRT_SECURE_NO_WARNINGS
* fp16 brain script - fix NetworkTests
* Update tests for model version change
* Remove changes for InputAndParamNodes
* Fix typo
* Remove redundant code
* Fix optional parameters
2019-03-20 11:36:16 -07:00
45ae386bc8
disable Manual/Tutorial tests on Windows Python2.7
2019-03-17 14:36:13 -07:00
9f60d8f815
Remove tab which caused build breaks
2019-03-15 10:47:16 -07:00
19ccd9f746
Address code review comments
2019-03-15 10:47:15 -07:00
3a21b0adbb
FP16 all reduce. Cast gradients to and from float16 before sending over NCCL to do all reduce.
2019-03-15 10:47:14 -07:00
4765f69c7a
support onnx export of 2GB models
2019-03-14 09:16:23 -07:00
22ded4679b
Address code review comments
2019-02-28 14:43:45 -08:00
d46ad9d7a3
Update
2019-02-28 14:43:45 -08:00
773379e622
Add more comments
2019-02-28 14:43:45 -08:00
d2dc1234c0
Fix BMUF fp16 to update local learner's fp32 cache
2019-02-28 14:43:45 -08:00
26e26e4078
Fix BMUF fp16 bug. Local learner ResetSmoothedGradients should NOT wipe cached fp32 model/parameters.
2019-02-28 14:43:44 -08:00
e104129369
Remove obsolete comments
2019-02-28 10:09:44 -08:00
92054e869a
Remove debug messages
2019-02-28 10:09:43 -08:00
730a2204f3
Add useV2Aggregator to SGD config
2019-02-28 10:09:43 -08:00
6ef0224aa5
Merge branch 'pull/3431'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-28 09:45:39 +00:00
39ca0a24ae
Merge branch 'pull/3482'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-28 09:04:17 +00:00
93f875cb49
Merge branch 'pull/3477'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-28 06:49:34 +00:00
8cbfbfec27
Merge branch 'pull/3458'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-28 04:38:53 +00:00
ef5b745587
Merge branch 'pull/3455'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-28 04:27:00 +00:00
1c5f124d28
Merge branch 'pull/3486'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-28 02:24:07 +00:00
726e48e536
Merge branch 'pull/3491'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-28 00:43:11 +00:00
771d754b8f
Merge branch 'pull/3492'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-27 23:59:43 +00:00
5ccf54a63d
Merge branch 'pull/3514'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-27 22:39:40 +00:00
af4287372a
Merge branch 'pull/3561'
...
Signed-off-by: Thiago Crepaldi <thiagofc@microsoft.com>
2019-02-26 21:07:30 +00:00
06469f7798
Merge branch 'pull/3558'
2019-02-25 18:46:26 +00:00
a5fb3d6f8d
add more onnx rnn model tests, enable optimizedRNN op test
2019-02-19 12:48:41 -08:00
00b80377fe
Update onnx submodule, support ConstantOfShape & OneHot
...
* Support SequenceIs[First,Last] with ConstantOfShape
* Update bypass load test in verify\_one\_input & add test for one hot
op
* Update export for one hot op. Migrate from exporting
onnx.ml.OneHotEncoder to onnx.OneHot.
Op fixes
* Fix topk onnx\_op\_test
* Support MVN export using ONNX function
* Fix LayerNormalization
* Skip tests for sequence slice float16: not supported in cntk
* Support gather export & import with float16
- Fix cntk gather issue with float16 inputs.
- Support exporting constant float16 tensor.
- Support importing int32 indices input for gather.
* Enable more passed op tests
2019-02-15 15:12:10 -08:00
aa828194db
Merge branch 'liqun/remove-crosstalkcaffe-test'
2019-02-14 21:07:34 +00:00
3bef0f4214
protobuf 3.6.0 to 3.6.1
2019-02-13 11:08:06 -08:00
9ce6c1c2f9
Merge pull request #3550 from Microsoft/yaronwe/hide-protobuf
...
hide protobuf library symbols
2019-02-08 08:37:36 -08:00
2ccdff45e3
enable onnx rnn test, name input with proper name
2019-02-03 21:18:50 -08:00
979c8bfe2a
Merge branch 'liqun/liqun/scan6Stage2'
2019-01-31 04:27:20 +00:00
9093a4def1
Cache CNTK dependencies at CNTK Azure Blob
2019-01-29 19:47:10 -08:00
6b2323c4e4
onnx support of sequential convolution
2019-01-28 14:48:13 -08:00
631ef0f9ed
Fix typo
...
Ture -> True
2019-01-29 00:02:51 +09:00
4ef44170ab
Adding sequence models to CI and skipping failing ones
2019-01-24 15:01:14 -08:00
fc839bc371
Moving CI test dependencies to python environments yml files
2019-01-22 16:24:29 -08:00
d67a5ae6cf
Free nnzPerRowOrCol at end of SetValue
2019-01-16 17:00:33 -05:00
fea6e9dd47
Merge branch 'vadimma/bmlfex'
2019-01-12 00:01:46 +00:00
62fa8182d7
Enable frame mode for the binary MLF reader
2019-01-10 18:02:34 -08:00
0be4bcb949
Merge branch 'liqun/scan5Stage'
2019-01-09 02:39:18 +00:00
fe939aa3cb
support ONNX scan(9)
2019-01-08 12:16:32 -08:00
a5fcc80d91
Resolve Linux binary drop & windows oobe test conda dependency
2019-01-07 15:07:04 -08:00
2274e024ed
hide protobuf library symbols
2019-01-06 10:04:34 +00:00
b732aeb0fd
Fix several DeprecationWarning: invalid escape sequence
...
Signed-off-by: Mickaël Schoentgen <contact@tiger-222.fr>
2019-01-05 22:36:45 +01:00
322293c8fd
Merge branch 'vadimma/adambs'
2019-01-04 00:21:27 +00:00
6c56fdd2f9
Revert "refactor"
...
This reverts commit 47345c6c54
2019-01-03 15:45:28 -08:00
47345c6c54
refactor
2019-01-03 15:26:03 -08:00
05435d0db2
Expose Adam in BS
2019-01-03 12:58:17 -08:00
08130dbb92
update linux binary drop cudnn path
2019-01-03 10:57:44 -08:00
rpengms
liqfu
Peggy Peng
Thiago Crepaldi
Bowen Bao
Amit Agarwal
Thiago Crepaldi
Takashi Takebayashi
Peyman Manikashani
Aaron Schweiger
Vadim Mazalov
yaron weinsberg
Mickaël Schoentgen