This is an automated email from the ASF dual-hosted git repository.
reminisce pushed a change to branch numpy
in repository https://gitbox.apache.org/repos/asf/incubator-mxnet.git.
discard bb3afd4 [Do not review] [Do not merge] New numpy-compatible sum
(#14739)
new 0da4b67 [MKLDNN]Add quantized relu (#14604)
new 18d4051 Add publish test of PyPi cu100mkl (#14637)
new 93238a2 [contrib][op] fix MultiBoxPrior confusing results if first
ratio is not 1.0 (#13763)
new 391a1be Set idx2name for Optimizer object (#14703)
new 818be02 [MXNET-1377] Add static-dependencies licenses (#14726)
new 5b6e25b [MXNET-1287] Feat dep (#14668)
new dc48cd2 License Googletest and Appendix (#14687)
new 153d2f4 [MXNET-1385] Improved Scala Init and Macros warning messages
(#14656)
new 5331933 Reenable TensorRT step (#14654)
new dd1004b fix pi instructions (#14746)
new 100586a Change size_t to int within for loop to fix windows build
error (#14740)
new 68efc15 fix custom op fork test (#14753)
new 494c29e [BUGFIX] fix ELU function will appear nan when calculating
the gradient (#14673)
new da7fff7 fix min max on zero-sized ndarray (#14745)
new a1b0a3a [Clojure] Better api docstrings by replacing newlines (#14752)
new 0f63659 add a compiler flag to use int64 as tensor size (#14570)
new 3b39c56 fix shape index bug (#14518)
new 2fd4720 fix acc_type_switch macro with extra tests (#14773)
new 2153b86 [Do not review] [Do not merge] New numpy-compatible sum
(#14739)
This update added new revisions after undoing existing revisions.
That is to say, some revisions that were in the old version of the
branch are not in the new version. This situation occurs
when a user --force pushes a change and generates a repository
containing something like this:
* -- * -- B -- O -- O -- O (bb3afd4)
\
N -- N -- N refs/heads/numpy (2153b86)
You should already have received notification emails for all of the O
revisions, and so the following emails describe only the N revisions
from the common base, B.
Any revisions marked "omit" are not gone; other references still
refer to them. Any revisions marked "discard" are gone forever.
The 9519 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
Summary of changes:
3rdparty/googletest | 2 +-
3rdparty/mshadow | 2 +-
CMakeLists.txt | 8 +
CONTRIBUTORS.md | 1 +
LICENSE | 42 ++-
Makefile | 5 +
ci/docker/runtime_functions.sh | 64 ++++-
ci/jenkins/Jenkins_steps.groovy | 43 ++-
ci/jenkins/Jenkinsfile_unix_cpu | 3 +-
ci/jenkins/Jenkinsfile_unix_gpu | 7 +-
contrib/clojure-package/src/dev/generator.clj | 16 +-
.../clojure-package/test/good-test-ndarray-api.clj | 85 +++++-
.../clojure-package/test/good-test-symbol-api.clj | 87 ++++++-
docs/install/index.md | 35 ++-
include/mxnet/libinfo.h | 6 +-
include/mxnet/tensor_blob.h | 13 +-
include/mxnet/tuple.h | 16 +-
make/config.mk | 6 +
make/crosscompile.jetson.mk | 6 +
make/osx.mk | 6 +
python/mxnet/gluon/nn/activations.py | 3 +-
python/mxnet/model.py | 2 +
python/mxnet/module/module.py | 16 +-
.../assembly/src/main/assembly/assembly.xml | 7 +-
.../scala/org/apache/mxnet/ExecutorManager.scala | 27 +-
.../main/scala/org/apache/mxnet/FeedForward.scala | 14 +-
.../core/src/main/scala/org/apache/mxnet/IO.scala | 35 ++-
.../src/main/scala/org/apache/mxnet/Image.scala | 8 +
.../scala/org/apache/mxnet/MX_PRIMITIVES.scala | 2 +
.../src/main/scala/org/apache/mxnet/NDArray.scala | 5 +
.../src/main/scala/org/apache/mxnet/Symbol.scala | 55 +++-
.../scala/org/apache/mxnet/io/MXDataIter.scala | 3 +-
.../scala/org/apache/mxnet/io/NDArrayIter.scala | 3 +-
.../org/apache/mxnet/io/PrefetchingIter.scala | 87 ++-----
.../scala/org/apache/mxnet/javaapi/Image.scala | 25 ++
.../scala/org/apache/mxnet/module/BaseModule.scala | 3 +-
.../org/apache/mxnet/module/BucketingModule.scala | 6 +-
.../scala/org/apache/mxnet/module/Module.scala | 4 +-
.../org/apache/mxnet/module/SequentialModule.scala | 16 +-
.../org/apache/mxnet/util/OptionConversion.scala | 2 +
.../src/test/scala/org/apache/mxnet/IOSuite.scala | 29 ++-
.../test/scala/org/apache/mxnet/ModuleSuite.scala | 4 +-
.../scala/org/apache/mxnet/train/ConvSuite.scala | 1 +
.../CNNTextClassification.scala | 2 +-
.../mxnetexamples/customop/ExampleCustomOp.scala | 2 +-
.../customop/ExampleCustomOpWithRtc.scala | 2 +-
.../datasets/SyntheticDataIter.scala | 2 +-
.../org/apache/mxnetexamples/module/MnistMlp.scala | 2 +-
.../mxnetexamples/module/SequentialModuleEx.scala | 2 +-
.../mxnetexamples/multitask/ExampleMultiTask.scala | 11 +-
.../mxnetexamples/neuralstyle/NeuralStyle.scala | 4 +-
.../neuralstyle/end2end/DataProcessing.scala | 4 +-
.../org/apache/mxnetexamples/rnn/BucketIo.scala | 4 +-
.../apache/mxnetexamples/rnn/TrainCharRnn.scala | 7 +-
.../benchmark/ScalaInferenceBenchmarkSuite.scala | 1 +
.../CNNClassifierExampleSuite.scala | 1 +
.../customop/CustomOpExampleSuite.scala | 1 +
.../apache/mxnetexamples/gan/GanExampleSuite.scala | 1 +
.../IMClassificationExampleSuite.scala | 1 +
.../ImageClassifierExampleSuite.scala | 1 +
.../ObjectDetectorExampleSuite.scala | 1 +
.../neuralstyle/NeuralStyleSuite.scala | 1 +
.../apache/mxnetexamples/rnn/ExampleRNNSuite.scala | 1 +
.../org/apache/mxnet/infer/ClassifierSuite.scala | 4 +-
.../main/scala/org/apache/mxnet/init/Base.scala | 27 +-
.../main/scala/org/apache/mxnet/init/LibInfo.scala | 30 +++
.../scala/org/apache/mxnet/APIDocGenerator.scala | 73 +++++-
.../scala/org/apache/mxnet/GeneratorBase.scala | 79 ++++--
.../main/scala/org/apache/mxnet/NDArrayMacro.scala | 61 ++++-
.../main/scala/org/apache/mxnet/SymbolMacro.scala | 58 ++++-
.../apache/mxnet/javaapi/JavaNDArrayMacro.scala | 15 +-
.../org/apache/mxnet/utils/CToScalaUtils.scala | 12 +-
.../apache/mxnet/utils/OperatorBuildUtils.scala | 32 ---
scala-package/pom.xml | 4 +
.../apache/mxnet/spark/io/LabeledPointIter.scala | 4 +-
.../mxnet/spark/io/LongLivingDataBatch.scala | 3 +-
.../org/apache/mxnet/spark/io/PointIter.scala | 4 +-
.../org/apache/mxnet/spark/MXNetGeneralSuite.scala | 1 +
src/common/serialization.h | 12 +-
src/imperative/cached_op.h | 8 +-
src/io/image_det_aug_default.cc | 4 +-
src/io/image_io.cc | 2 +-
src/io/iter_batchloader.h | 2 +-
src/io/iter_sparse_batchloader.h | 2 +-
src/libinfo.cc | 3 +
src/operator/contrib/dgl_graph.cc | 4 +-
src/operator/contrib/multi_proposal-inl.h | 16 +-
src/operator/contrib/multibox_detection-inl.h | 2 +-
src/operator/contrib/multibox_detection.cc | 2 +-
src/operator/contrib/multibox_detection.cu | 2 +-
src/operator/contrib/multibox_prior-inl.h | 8 +-
src/operator/contrib/multibox_prior.cc | 7 +-
src/operator/contrib/multibox_prior.cu | 5 +-
src/operator/contrib/multibox_target-inl.h | 2 +-
src/operator/contrib/multibox_target.cc | 2 +-
src/operator/contrib/multibox_target.cu | 2 +-
src/operator/contrib/proposal-inl.h | 16 +-
src/operator/convolution_v1-inl.h | 2 +-
src/operator/image/image_random-inl.h | 14 +-
src/operator/mxnet_op.h | 4 +
src/operator/nn/deconvolution-inl.h | 2 +-
src/operator/nn/mkldnn/mkldnn_act-inl.h | 74 ++++++
src/operator/nn/mkldnn/mkldnn_act.cc | 91 +++----
src/operator/optimizer_op-inl.h | 8 +-
...uantized_pooling.cc => mkldnn_quantized_act.cc} | 38 +--
.../mkldnn/mkldnn_quantized_fully_connected.cc | 2 +-
src/operator/quantization/quantize_graph_pass.cc | 29 ++-
src/operator/quantization/quantized_activation.cc | 138 ++++++++++
src/operator/subgraph/mkldnn/mkldnn_fc.cc | 2 +-
src/operator/swapaxis-inl.h | 4 +-
src/operator/tensor/broadcast_reduce-inl.cuh | 13 +
src/operator/tensor/broadcast_reduce-inl.h | 13 +
src/operator/tensor/broadcast_reduce_op.h | 27 ++
src/operator/tensor/broadcast_reduce_op_value.cc | 4 +-
src/operator/tensor/indexing_op.h | 4 +-
src/operator/tensor/matrix_op-inl.h | 28 +-
src/operator/tensor/sparse_retain-inl.h | 2 +-
src/operator/tensor/square_sum-inl.h | 9 +-
tests/nightly/JenkinsfileForBinaries | 35 +++
.../nightly/apache_rat_license_check/rat-excludes | 1 +
tests/nightly/test_large_array.py | 28 +-
tests/python/gpu/test_operator_gpu.py | 1 +
tests/python/quantization/test_quantization.py | 55 +++-
tests/python/unittest/test_contrib_operator.py | 14 +-
tests/python/unittest/test_gluon.py | 2 +-
tests/python/unittest/test_module.py | 28 ++
tests/python/unittest/test_operator.py | 55 +++-
tools/dependencies/LICENSE.binary.dependencies | 289 +++++++++++++++++++++
tools/dependencies/eigen.sh | 1 +
tools/staticbuild/build.sh | 7 +
tools/staticbuild/build_lib.sh | 7 -
131 files changed, 1916 insertions(+), 484 deletions(-)
delete mode 100644
scala-package/macros/src/main/scala/org/apache/mxnet/utils/OperatorBuildUtils.scala
create mode 100644 src/operator/nn/mkldnn/mkldnn_act-inl.h
copy src/operator/quantization/mkldnn/{mkldnn_quantized_pooling.cc =>
mkldnn_quantized_act.cc} (54%)
create mode 100644 src/operator/quantization/quantized_activation.cc
create mode 100644 tools/dependencies/LICENSE.binary.dependencies
