This is an automated email from the ASF dual-hosted git repository.
lausen pushed a change to branch leezu-patch-1
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
from 254a4cd Update to Intel(R) Math Kernel Library 2020 Update 4
add 33d94f1 Enable Numpy support for Gluon Block optimize_for (#19455)
add ea222a3 Update README.md (#19438)
add b33fbd1 Make RTC compatible with CUDA enhanced compatibility (#19364)
add 3d1df4e Support both Gluon 1 and 2 in the hybrid containers (#19470)
add aa4e3ac Mark test_lstm_clip as flaky (#19474)
add 07cd205 [FIX] CheckpointHandler verbose behavior in
gluon.contrib.estimator.event_handler (#19471)
add 087f6ff [BUGFIX] Fallback to native implementation of RNN when
use_sequence_length=True (#19466)
add 27ae77d numpy tril_indices large tensor fix (#19381)
add 94f55b1 Avoid segfault when calling __repr__ on ndarray or symbol
with freed handle (#19475)
add b409371 Skip flaky memory profiler tests (#19346)
add c0c4cb5 Update LICENSE (#19478)
add a163c2b Skip flaky segfaulting test_bbox_augmenters on Windows
(#19480)
add 31d6151 Mark test_sparse_storage_fallback as flaky (#19482)
add 77c4dbb Check if backend memory has been freed in check_leak_ndarray
pytest fixture (#19483)
add f3c69c1 Add test case for oneDNN RNN (#19465)
add d7fd787 [FEATURE] Enable large tensor support for insert (#19340)
add 2fa99b2 Fix a typo on dataset.py (#19452)
add 63d846c Allow eliminating common subexpressions when temp space is
used (#19486)
add abce30a Mark test_random_gray as flaky (#19493)
add 545e117 Mark test_batchnorm_with_type as flaky (#19492)
add d4a5f2d Add SiLU (#19497)
add 564c6d3 Extension bug fixes (#19469)
add 551a8d3 Update pytest version used on CI (#19491)
add d46dc96 Disable ENABLE_TESTCOVERAGE on CentOS 7 build (#19507)
add 74e5392 Unique names for nodes in imperative mode (#19509)
add f79e5bd Update CONTRIBUTORS.md (#19512)
add a69e2b4 Enable coverage reports for PR diffs (#19505)
add 495a197 Mark test_gluon_bernoulli_v1 as flaky (#19510)
add ced029a Mark test_npx_batch_norm as flaky (#19508)
add 1120d59 Fix markdown syntax for link to community guide (#19513)
add a52b639 Optimize graph execution in presence of dynamic shape op
(#18690)
add b3a6317 Fix HLS2RGBConvert (#19516)
add b9e8efa Update CONTRIBUTORS.md (#19518)
add ea4e2b3 fix msvc C2398 error (#19500)
add abeaba7 disable graph partition by default (#19525)
add 4da8783 Fix uninitialized variable in RTC path (#19528)
add d43f439 Enable Numpy Large tensor nightly tests (#19529)
add bb8f6ce change partitioning default to False for hybridize and
optimize_for (#19531)
add 3bf556d Cuda 11 build fixes (#19530)
add 6bc0647 External Operators 2 (#19431)
add 9625f8f Fix fallback.py and test_numpy_interoperability.py (#19457)
add 5000ccb Update pip install instruction (#19549)
add fcfef81 Enable Large Tensor Support by default (#18625)
add ab7ad4d Add AdamW optimizer (#19526)
add 07c1e1e Don't declare/use CheckPvalGPU on CPU builds (#19545)
add 18a784a Numpy take large tensor fix (#19325)
add d2818b4 CPU reduce kernel optimization (#19419)
add f998ac6 Mark test_np_average as flaky (#19538)
add 6fa208b Fix compilation for large tensor with MKL (#19067)
add dbcd6a1 Separate backend from hybridize and refactor optimize_for
kwargs (#19543)
add c9d9c85 Mark test_np_around as flaky (#19554)
add 5dc404d NumPy compatible serialization API (#19417)
add 6648866 Add AMP patching of npi ops in _api_internal module (#19488)
add 13e9d57 Fix compilation with USE_LAPACK=0 USE_BLAS=mkl (#19570)
add 221344e Convert symbol to numpy symbol in Symbol class (#19523)
add 78a0f7b [submodule] Upgrade to oneDNN v1.7 (#19558)
add f97544e Update ONNX support to 1.7 (#19573)
add 16be29d Mark test_dropout as flaky (#19553)
add d599a15 Update cmake/BuildTVM.cmake to use system llvm (#19589)
add 36b54b9 Switch npx.savez from libzip to miniz (#19584)
add fef85c1 Remove einsum overhead (#19421)
add 748e100 [FEATURE] Add BN+ReLU -> BatchNormWithRelu fuse to MKLDNN
backend (#19578)
add 2db601a Mark test_cv_model_inference_onnxruntime_mobilenetv2_1_0 as
flaky (#19590)
add 07bfe55 parametrize test_deformable_convolution (#19600)
add 78fc40f Fix serialization bug for writing large arrays to npz (#19596)
add 9a1a27f Merge remote-tracking branch 'origin/master'
add ed16d9b RTLD_LOCAL -> RTLD_GLOBAL
No new revisions were added by this update.
Summary of changes:
.codecov.yml | 4 +-
.github/workflows/os_x_staticbuild.yml | 6 +-
.gitmodules | 3 +
3rdparty/miniz/miniz.c | 7657 ++++++++++++++++++++
3rdparty/miniz/miniz.h | 1365 ++++
3rdparty/mkldnn | 2 +-
3rdparty/mshadow/mshadow/base.h | 16 +
3rdparty/mshadow/mshadow/dot_engine-inl.h | 28 +-
3rdparty/mshadow/mshadow/tensor.h | 11 +
CMakeLists.txt | 21 +-
CONTRIBUTORS.md | 5 +-
LICENSE | 876 +--
NOTICE | 11 +
README.md | 2 +-
ci/build_windows.py | 19 +-
ci/docker/Dockerfile.build.android | 26 +-
ci/docker/Dockerfile.build.arm | 30 +-
ci/docker/Dockerfile.build.jetson | 2 +-
ci/docker/Dockerfile.build.ubuntu | 3 +-
ci/docker/Dockerfile.test.arm | 8 +-
ci/docker/install/requirements | 17 +-
ci/docker/runtime_functions.sh | 55 +-
ci/jenkins/Jenkins_steps.groovy | 14 +-
cmake/BuildTVM.cmake | 5 +-
cmake/ChooseBlas.cmake | 8 +
config/darwin.cmake | 2 +-
config/linux.cmake | 2 +-
config/linux_gpu.cmake | 2 +-
contrib/tvmop/prepare_tvm.sh | 63 -
docker/install/python.sh | 2 +-
docs/python_docs/python/api/gluon/nn/index.rst | 2 +
.../gluon/blocks/activations/activations.md | 34 +-
.../gluon/blocks/activations/images/gelu.png | Bin 0 -> 10289 bytes
.../gluon/blocks/activations/images/silu.png | Bin 0 -> 10155 bytes
.../python/tutorials/packages/np/np-vs-numpy.md | 9 +-
.../_includes/get_started/linux/python/gpu/pip.md | 2 +-
docs/static_site/src/assets/js/clipboard.js | 24 +-
example/extensions/lib_external_ops/CMakeLists.txt | 18 +
example/extensions/lib_external_ops/README.md | 70 +
.../{lib_api => lib_external_ops}/init_lib.cc | 4 +-
example/extensions/lib_external_ops/min_ex-inl.h | 66 +
.../extensions/lib_external_ops/min_ex.cc | 22 +-
.../extensions/lib_external_ops/min_ex.cu | 13 +-
.../{lib_api => lib_external_ops}/test_loading.py | 27 +-
example/extensions/lib_pass/README.md | 17 +-
example/extensions/lib_pass/test_pass.py | 10 -
example/extensions/lib_subgraph/README.md | 38 +-
example/extensions/lib_subgraph/test_subgraph.py | 18 +-
include/mxnet/base.h | 2 +
include/mxnet/c_api.h | 20 +
include/mxnet/rtc.h | 2 +-
licenses/BOOST1_0 | 23 +
licenses/BSD2 | 7 +
licenses/BSD3-cmake | 30 +
.../themes/mx-theme/LICENSE => licenses/MIT | 12 +-
licenses/OFL1_1 | 97 +
python/mxnet/_ctypes/ndarray.py | 4 +-
python/mxnet/_ctypes/symbol.py | 4 +-
python/mxnet/amp/amp.py | 46 +-
python/mxnet/base.py | 6 +-
.../mxnet/contrib/onnx/mx2onnx/_op_translations.py | 217 +-
python/mxnet/contrib/onnx/mx2onnx/export_model.py | 10 +-
python/mxnet/contrib/onnx/mx2onnx/export_onnx.py | 11 +-
.../mxnet/contrib/onnx/onnx2mx/_op_translations.py | 111 +-
python/mxnet/contrib/onnx/onnx2mx/import_model.py | 3 +-
python/mxnet/contrib/onnx/onnx2mx/import_onnx.py | 8 +-
.../mxnet/contrib/onnx/onnx2mx/import_to_gluon.py | 3 +-
python/mxnet/cython/ndarray.pyx | 3 +
python/mxnet/cython/symbol.pyx | 3 +
python/mxnet/gluon/block.py | 174 +-
.../mxnet/gluon/contrib/estimator/event_handler.py | 2 +-
python/mxnet/gluon/data/dataset.py | 2 +-
python/mxnet/gluon/nn/activations.py | 33 +-
python/mxnet/gluon/nn/basic_layers.py | 43 +-
python/mxnet/ndarray/ndarray.py | 11 +-
python/mxnet/ndarray/utils.py | 11 +-
python/mxnet/numpy/fallback.py | 14 +-
python/mxnet/numpy/multiarray.py | 50 +-
python/mxnet/numpy_extension/utils.py | 122 +-
python/mxnet/optimizer/__init__.py | 4 +-
python/mxnet/optimizer/adamW.py | 233 +
python/mxnet/symbol/numpy/_symbol.py | 19 +-
python/mxnet/symbol/symbol.py | 22 +-
src/api/operator/numpy/np_delete_op.cc | 16 +-
src/api/operator/numpy/np_insert_op.cc | 12 +-
src/api/operator/numpy/np_matrix_op.cc | 12 +-
src/c_api/c_api.cc | 199 +-
src/c_api/c_api_symbolic.cc | 20 +
src/common/cuda/rtc.cc | 48 +-
src/common/cuda/rtc/half-inl.h | 2 +-
src/common/cuda/rtc/vectorization-inl.h | 4 +-
src/common/rtc.cc | 37 +-
src/imperative/eliminate_common_expr_pass.cc | 10 +-
src/imperative/imperative.cc | 6 +-
src/imperative/imperative_utils.h | 10 +-
src/operator/c_lapack_api.h | 69 +-
src/operator/contrib/batch_norm_relu.cc | 11 +-
src/operator/contrib/multi_lamb.cu | 8 +-
src/operator/contrib/multi_lans.cu | 8 +-
src/operator/contrib/transformer.cc | 14 +-
src/operator/control_flow.cc | 2 +-
src/operator/fusion/fused_op.cu | 2 +-
src/operator/image/image_random-inl.h | 3 +-
src/operator/linalg.h | 24 +-
src/operator/linalg_impl.h | 88 +-
src/operator/nn/mkldnn/mkldnn_rnn-inl.h | 5 +
src/operator/numpy/linalg/np_gesvd-inl.h | 2 +-
src/operator/numpy/linalg/np_lstsq-inl.h | 185 +-
src/operator/numpy/linalg/np_lstsq.cc | 20 +-
src/operator/numpy/linalg/np_matrix_rank-inl.h | 7 +-
src/operator/numpy/linalg/np_pinv-inl.h | 66 +-
src/operator/numpy/linalg/np_solve-inl.h | 75 +-
src/operator/numpy/linalg/np_tensorsolve-inl.h | 45 +-
src/operator/numpy/np_delete_op-inl.h | 52 +-
src/operator/numpy/np_einsum_op-inl.h | 3 +-
src/operator/numpy/np_einsum_op.cu | 122 +-
src/operator/numpy/np_insert_op-inl.h | 82 +-
src/operator/numpy/np_insert_op_scalar-inl.h | 8 +-
src/operator/numpy/np_insert_op_slice-inl.h | 25 +-
src/operator/numpy/np_insert_op_slice.cc | 10 +-
src/operator/numpy/np_insert_op_tensor-inl.h | 27 +-
src/operator/numpy/np_insert_op_tensor.cc | 8 +-
src/operator/numpy/np_matrix_op-inl.h | 36 +-
src/operator/numpy/np_matrix_op.cc | 23 +-
src/operator/numpy/np_polynomial_op.cc | 3 +-
src/operator/numpy/random/np_multinomial_op.h | 11 +
src/operator/rnn.cc | 6 +-
src/operator/subgraph/build_subgraph.cc | 54 +-
.../subgraph/mkldnn/mkldnn_bn_relu_property.h | 142 +
.../subgraph/mkldnn/mkldnn_subgraph_property.cc | 21 +-
.../partitioner/custom_subgraph_property.h | 74 +-
.../subgraph/static_shape_subgraph_property.cc | 189 +
src/operator/subgraph_op_common.cc | 4 +-
src/operator/subgraph_op_common.h | 12 +-
src/operator/tensor/broadcast_reduce-inl.h | 59 +-
src/operator/tensor/indexing_op.cc | 2 +-
src/operator/tensor/indexing_op.h | 20 +-
src/operator/tensor/la_op-inl.h | 87 +-
src/operator/tensor/la_op.cc | 2 +
src/operator/tensor/la_op.h | 36 +-
src/serialization/cnpy.cc | 896 +++
.../api_registry.h => src/serialization/cnpy.h | 47 +-
tests/cpp/operator/mkldnn_test.cc | 2 +-
tests/nightly/JenkinsfileForBinaries | 1 -
tests/nightly/test_np_large_array.py | 57 +-
tests/python/conftest.py | 4 +-
.../{mkl/test_subgraph.py => gpu/test_amp_init.py} | 50 +-
tests/python/gpu/test_extensions_gpu.py | 25 +-
tests/python/gpu/test_gluon_gpu.py | 2 +-
tests/python/gpu/test_operator_gpu.py | 4 +-
tests/python/gpu/test_profiler_gpu.py | 2 +-
tests/python/mkl/test_mkldnn.py | 28 +
tests/python/mkl/test_subgraph.py | 18 +
tests/python/unittest/onnx/backend.py | 15 +-
tests/python/unittest/onnx/test_onnxruntime.py | 182 +
tests/python/unittest/test_deferred_compute.py | 17 +
tests/python/unittest/test_dynamic_shape.py | 75 +
tests/python/unittest/test_extensions.py | 47 +-
tests/python/unittest/test_gluon.py | 81 +-
tests/python/unittest/test_gluon_data_vision.py | 2 +-
tests/python/unittest/test_gluon_probability_v1.py | 1 +
tests/python/unittest/test_ndarray.py | 45 +-
.../test_numpy_contrib_gluon_data_vision.py | 2 +
tests/python/unittest/test_numpy_gluon.py | 30 +-
.../unittest/test_numpy_gluon_data_vision.py | 1 +
.../python/unittest/test_numpy_interoperability.py | 4 +-
tests/python/unittest/test_numpy_ndarray.py | 66 +-
tests/python/unittest/test_numpy_op.py | 3 +
tests/python/unittest/test_operator.py | 118 +-
tests/python/unittest/test_optimizer.py | 25 +
tests/python/unittest/test_profiler.py | 1 +
tests/python/unittest/test_sparse_ndarray.py | 121 +-
tests/python/unittest/test_sparse_operator.py | 1 +
tests/python/unittest/test_subgraph_op.py | 19 +-
tools/license_header.py | 3 -
175 files changed, 13859 insertions(+), 2265 deletions(-)
create mode 100644 3rdparty/miniz/miniz.c
create mode 100644 3rdparty/miniz/miniz.h
delete mode 100644 contrib/tvmop/prepare_tvm.sh
create mode 100644
docs/python_docs/python/tutorials/packages/gluon/blocks/activations/images/gelu.png
create mode 100644
docs/python_docs/python/tutorials/packages/gluon/blocks/activations/images/silu.png
create mode 100644 example/extensions/lib_external_ops/CMakeLists.txt
create mode 100644 example/extensions/lib_external_ops/README.md
copy example/extensions/{lib_api => lib_external_ops}/init_lib.cc (92%)
create mode 100644 example/extensions/lib_external_ops/min_ex-inl.h
copy src/operator/numpy/np_triu_op.cu =>
example/extensions/lib_external_ops/min_ex.cc (61%)
copy src/operator/numpy/np_interp_op.cu =>
example/extensions/lib_external_ops/min_ex.cu (80%)
copy example/extensions/{lib_api => lib_external_ops}/test_loading.py (71%)
create mode 100644 licenses/BOOST1_0
create mode 100644 licenses/BSD2
create mode 100644 licenses/BSD3-cmake
copy docs/python_docs/themes/mx-theme/LICENSE => licenses/MIT (87%)
create mode 100644 licenses/OFL1_1
create mode 100644 python/mxnet/optimizer/adamW.py
create mode 100644 src/operator/subgraph/mkldnn/mkldnn_bn_relu_property.h
create mode 100644 src/operator/subgraph/static_shape_subgraph_property.cc
create mode 100644 src/serialization/cnpy.cc
copy include/mxnet/api_registry.h => src/serialization/cnpy.h (57%)
copy tests/python/{mkl/test_subgraph.py => gpu/test_amp_init.py} (51%)
create mode 100644 tests/python/unittest/onnx/test_onnxruntime.py
