82d350d0ac
bump up version number
2018-09-13 15:52:47 -07:00
deda94b67b
Support pooling(cpu) where kernel center is on pads.
...
- Previous implementation has the assumption that (0 <= dk < width).
This assumption doesn't stand when lo >(kernel - 1) / 2.
The updated calculation supports arbitrary lo & hi non-negative
integer value. The new calculation has dk in range (0, width + hi +
lo].
- Enables onnx backend test {averagepool_2d_pads, maxpool_2d_pads} to
pass.
2018-09-12 21:37:21 -07:00
61d7dab912
Support more than 2 inputs for ONNX Min/Max import.
2018-09-12 15:12:14 -07:00
61572e89f8
Update onnx_model_test skip list
2018-09-11 08:51:45 -07:00
0754b38e34
update onnx_model_test with tests from onnx backend test
2018-09-09 13:53:26 -07:00
d877233979
Make broadcast ops compitable between CNTK and ONNX,
...
Enable ONNX export/import for optimizedRNN op,
More ONNX support for Sequence ops
2018-09-09 08:59:33 -07:00
fcf9f48895
Overhaul conv/convTrans/pooling pads value export
...
- Update exporting of conv/pooling to always export pad values.
- Enable correct exporting of multiple pretrained models (ResNet50/ResNet101/ResNet152_ImageNet_Caffe, etc).
- Overhaul convtranspose pads exporting
- Support conv weight export with omitted out channel axis (LRN).
- Add tests in onnx_op_test to cover the above changes
2018-09-06 11:46:14 -07:00
e3a1acfdf0
Resolve dependencies and build issues
...
-Temporary add importorskip around import onnx
-bump up .yml matplotlib version
2018-09-05 15:02:23 -07:00
77a8c4992f
Temporarily skip onnx_model_test if import onnx fail
2018-08-30 10:52:53 -07:00
18d9f39afc
skip dynamic axes wrapper, export onnx test cases, handle output op being Combine op, workaround a specific case bug of ONNX bidirectional broadcast shape inference
2018-08-29 16:52:58 -07:00
b3c0fa2c6b
Overhaul ConvTranpose to match ONNX 1.2.2. spec.
2018-08-26 21:20:05 -07:00
7dd9638799
squash of the following changes:
...
- fix flatten onnx export.
- fix unsqueeze onnx export.
- add comments on temporarily skipped tests.
- adjust the importing of softmax, logsoftmax and hardmax with blockfunction
- such that they could be exported as is back to onnx.
- update reshape onnx export to pass mobilenet round trip test.
2018-08-26 13:11:44 -07:00
0e208365be
CNTK splice allows broadcast. This case is handled in the change.
For noop (identity) ops, its inputs and outputs types shall be set according to upstream ops.
ToBatch/ToSequence and Unpack Batch/Sequence ops added during model importing need tp be skipped.
Model import need to handle ops with multiple outputs
2018-08-26 08:41:20 -07:00
a26e542c88
update conftest.py to resolve doctest issue.
...
- newer version numpy has a different print format for arrays and scalars that would potentially break the doctests.
2018-08-23 21:28:06 -07:00
28ada9657b
update python doctest handling newer version numpy print format.
2018-08-22 23:21:13 -07:00
56ef694c88
add onnx_model_test.py
2018-08-22 23:21:13 -07:00
4244320eba
Adding support for exporting CNTK TimesTranspose
2018-08-22 11:33:55 -07:00
8a244e6b3e
fix onnx_op_test.py reduce ops.
...
- remove bracket around data to match latest verification.
2018-08-21 14:23:28 -07:00
db41aebbad
Merge branch 'bowbao/onnx_reduce_allaxes'
2018-08-21 10:05:14 -07:00
4bbb245cc9
Merge branch 'liqun/TopKStage'
2018-08-21 03:47:21 +00:00
0e57d20b2c
fix reduce ops for onnx export/import with all_static_axes() and
...
all_axes().
2018-08-20 14:48:42 -07:00
3328f82a58
update onnx_backend_test.py skipped models and ops.
2018-08-20 11:57:52 -07:00
1a572f89f0
ONNX support of TopK
2018-08-17 13:01:47 -07:00
4b2b0d30cc
Merge branch 'peykash/batchnorm_float16_fixes'
2018-08-17 09:54:13 +00:00
490620873f
Merge branch 'bowbao/onnx_imageScaler'
2018-08-15 18:01:32 -07:00
bff3002794
fix python 2.7 tuple unpacking error
2018-08-15 13:47:56 -07:00
4b4b9cda28
fix imageScaler export/import to onnx
2018-08-15 10:22:47 -07:00
a5421862de
BatchNormalization Float16 Fixes
2018-08-15 10:01:23 -07:00
ef836bc293
Merge branch 'liqun/NewSequenceSliceStage2Stage'
2018-08-15 06:46:32 +00:00
6547e2ce7f
Fix bug in import of Conv op in ONNX.
2018-08-14 16:00:38 -07:00
ae163b33a8
ONNX support for CNTK Sequence::Slice op. Due to a specific feature of CNTK Sequence::Slice op to handle zigged inputs (batch of variable length sequences), models converted from CNTK may not produce marching outcomes with zigged inputs.
2018-08-14 11:17:36 -07:00
541e2100eb
Merge branch 'master' into sptiwari/model_fix_test
2018-08-13 10:48:20 -07:00
b689ebb5ad
Ignore TEST_TAG environment variable when not defined during testing
2018-08-13 09:02:01 -07:00
4ce6607917
Disable fmeasure test and increase pytest verbosity
2018-08-12 13:24:20 -07:00
e0b26561b8
Fixing batch axis handling bug in ReduceElements and Reshape export.
2018-08-10 16:36:29 -07:00
ff118cc684
Merge branch 'master' into sptiwari/onnx_merge_with_master
2018-08-10 10:10:06 -07:00
b7811d3549
Fmeasure ( #3343 )
...
* amended a bug, swapped precision and recall
* added a bit more documentation
2018-08-10 08:50:42 -07:00
a87d69a778
Fix Hardmax/Softmax/LogSoftmax ONNX export.
2018-08-08 16:52:33 -07:00
5e871a19fe
treat sequence axis as static axis for ONNX RNN model import
2018-08-02 10:51:05 -07:00
6f7ceed3bf
Merge branch 'pull/3284'
2018-08-01 04:48:10 +00:00
279a466b5f
Squash of the following
...
1. fix gather and prelu arguments order issue.
2. add typenameToTypeProto map initialization.
3. add CNTK native default lotus logger.
4. add CNTK unsqueeze op.
2018-07-30 19:58:55 -07:00
3d753db8a3
Enabled set random seed function and force deterministic api in c#
2018-07-30 16:06:34 -07:00
2964f1fde5
batchnorm_without_batch_axis_fixes
2018-07-30 12:42:35 -07:00
b7f7b5360f
Update Flatten/Reshape ONNX conversions for Softmax/Hardmax/LogSoftmax.
2018-07-27 14:43:52 -07:00
6783e9ce96
Fix Hardmax/Softmax/LogSoftmax ONNX import/export.
2018-07-27 14:37:16 -07:00
8a219b18c9
Don't crash on non-constant shape on ONNX Reshape import.
2018-07-25 17:58:12 -07:00
372aea564c
Changed the default Dispose method generated by Swig for Value class to call
...
Value.Erase before deleting the pointer in unmanaged code.
2018-07-24 11:37:31 -07:00
50d62a67d6
Merge branch 'sedymche/onnx-backend-test' into cntkteam/onnx_without_batch_axis
2018-07-24 03:49:04 +00:00
4a536b5b3b
Enabling more round trip tests - mean, sum, select.
2018-07-23 17:31:24 -07:00
afd542f277
Update ONNX backend test skip list.
2018-07-23 17:06:20 -07:00
liqfu
Bowen Bao
Sergii Dymchenko
Spandan Tiwari
Peyman Manikashani
Thiago Crepaldi
Thiago Crepaldi
delzac
Ke Deng
TJ