You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@mxnet.apache.org by ha...@apache.org on 2019/12/16 06:39:40 UTC

[incubator-mxnet] branch master updated (3d38dbd -> 7dbb4b7)

This is an automated email from the ASF dual-hosted git repository.

haoj pushed a change to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.


    from 3d38dbd  Adds min cuda version assertion decorator for unit tests (#17054)
     add 7dbb4b7  fix precision problem in linalg_solve, linalg_tensorinv, linalg_cholesky op test (#16981)

No new revisions were added by this update.

Summary of changes:
 python/mxnet/ndarray/numpy/linalg.py               |  57 ++++++-
 python/mxnet/numpy/linalg.py                       |  57 ++++++-
 python/mxnet/numpy_dispatch_protocol.py            |   1 +
 python/mxnet/symbol/numpy/linalg.py                |  57 ++++++-
 src/operator/numpy/linalg/np_tensorinv-inl.h       | 166 +++++++++++++++++++++
 src/operator/numpy/linalg/np_tensorinv.cc          | 134 +++++++++++++++++
 .../numpy/linalg/{np_gesvd.cu => np_tensorinv.cu}  |  14 +-
 .../python/unittest/test_numpy_interoperability.py |  26 ++++
 tests/python/unittest/test_numpy_op.py             | 166 ++++++++++++++++++---
 9 files changed, 646 insertions(+), 32 deletions(-)
 create mode 100644 src/operator/numpy/linalg/np_tensorinv-inl.h
 create mode 100644 src/operator/numpy/linalg/np_tensorinv.cc
 copy src/operator/numpy/linalg/{np_gesvd.cu => np_tensorinv.cu} (75%)