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 f01dc80 Adding sparse support to MXTensor for custom operators
(#17569)
add 2f358fd [Numpy] Add op fmax, fmin, fmod (#17567)
No new revisions were added by this update.
Summary of changes:
benchmark/python/ffi/benchmark_ffi.py | 3 +
python/mxnet/ndarray/numpy/_op.py | 76 ++++++++++-
python/mxnet/numpy/multiarray.py | 96 +++++++++++++-
python/mxnet/numpy_dispatch_protocol.py | 3 +
python/mxnet/symbol/numpy/_symbol.py | 24 +++-
.../np_elemwise_broadcast_op_extended_sec.cc} | 52 ++++----
src/operator/mshadow_op.h | 50 ++++++++
.../numpy/np_elemwise_broadcast_op_extended_sec.cc | 142 +++++++++++++++++++++
.../numpy/np_elemwise_broadcast_op_extended_sec.cu | 77 +++++++++++
src/operator/operator_tune.cc | 4 +
.../python/unittest/test_numpy_interoperability.py | 24 ++++
tests/python/unittest/test_numpy_op.py | 10 ++
12 files changed, 525 insertions(+), 36 deletions(-)
copy src/api/{_api_internal/_api_internal.cc =>
operator/numpy/np_elemwise_broadcast_op_extended_sec.cc} (55%)
create mode 100644 src/operator/numpy/np_elemwise_broadcast_op_extended_sec.cc
create mode 100644 src/operator/numpy/np_elemwise_broadcast_op_extended_sec.cu
