This is an automated email from the ASF dual-hosted git repository. taolv pushed a change to branch mkldnn-v1.0 in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from b5eb25d disable MKLDNN FC backward add 86ed5f5 [NumPy][Operator] NumPy operator `may_share_memory` and `shares_memory` (#16533) add 60d74bc Showing proper error message when an attempt is made to create large tensor but MXNet is not built with it (#16570) add 5aa74e0 Move ops which don't support FP16 dtype to FP32 list (#16668) add 8e50fd9 no such method => modified function args (#16610) add 27bddf8 [Numpy] Numpy operator diff (#15906) add 109be60 Merge remote-tracking branch 'origin/master' into mkldnn-v1.0 No new revisions were added by this update. Summary of changes: include/mxnet/c_api.h | 12 +- julia/deps/build.jl | 2 +- julia/src/metric.jl | 2 +- python/mxnet/contrib/amp/lists/symbol.py | 6 +- python/mxnet/ndarray/ndarray.py | 10 + python/mxnet/ndarray/numpy/_op.py | 124 +++++++++++- python/mxnet/numpy/multiarray.py | 128 +++++++++++- python/mxnet/numpy_dispatch_protocol.py | 5 +- python/mxnet/symbol/numpy/_symbol.py | 90 ++++++++- python/mxnet/symbol/symbol.py | 7 +- src/c_api/c_api.cc | 13 +- src/c_api/c_api_ndarray.cc | 8 +- src/ndarray/ndarray.cc | 15 ++ src/ndarray/ndarray_function.cc | 5 + src/operator/numpy/np_diff-inl.h | 220 +++++++++++++++++++++ src/operator/numpy/np_diff.cc | 109 ++++++++++ .../numpy/{random/np_uniform_op.cu => np_diff.cu} | 14 +- src/operator/numpy/np_memory_op.cc | 62 ++++++ .../quantize.cu => numpy/np_memory_op.cu} | 12 +- src/operator/numpy/np_memory_op.h | 75 +++++++ src/operator/tensor/init_op.h | 30 ++- .../python/unittest/test_numpy_interoperability.py | 39 ++++ tests/python/unittest/test_numpy_op.py | 79 +++++++- tests/python/unittest/test_operator.py | 56 ++++++ 24 files changed, 1087 insertions(+), 36 deletions(-) create mode 100644 src/operator/numpy/np_diff-inl.h create mode 100644 src/operator/numpy/np_diff.cc copy src/operator/numpy/{random/np_uniform_op.cu => np_diff.cu} (75%) create mode 100644 src/operator/numpy/np_memory_op.cc copy src/operator/{quantization/quantize.cu => numpy/np_memory_op.cu} (82%) create mode 100644 src/operator/numpy/np_memory_op.h