* add INT8 conv+relu6 fuse and enbale mobilentv2 INT8 test
test=develop
* change fasle and 0.0 to fuse_brelu and brelu_threshold
test=develop
change the "fuse_relu||fuse_brelu" to "unsigned_output"
test=develop
* Use relu instead of brelu as INT8 post-op because INT8 brelu is not enabled in mkldnn v0.18
test=develop
* continuous-integration fix
test=develop
* add Concat quantization
add unit test for quantizing concat
fix for wrong value when the input is not in map of calculated scales
add use_quantizer to concat_op.cc
add scale_algo rules for concat
test=develop
* missing fix for multiple inputs quantize-squash
* wojtuss review fix: adding comment
test=develop
* fix the bug that sub_scope_ may be null in AnalysisPredictor::Run.
* add more directions about io APIs' docs.
* update the API.spec. test=develop test=document_preview
* fluid int8 train and trt int8 predict align.
trt int8 predict init
op converter
* 2. align fluid int8 train and trt int8 inference.
enhance quant dequant fuse pass
enhance op converter, trt engine, trt engine op, trt subgraph pass.
* 3. add delete_quant_dequant_pass for trt
test=develop
* 4. add the missing file
test=develop
* 5. i modify the c++ interface, but forget to modify the pybind code
fix the IS_TRT_VERSION_GE bug, and fix elementwise op converter
test=develop
* fuse mul and elementwise add to fc
* Reimplement the FC forward operator
* Fix FC MKLDNN integration by transposing weights
* Add FC MKLDNN Pass
test=develop
* FC MKLDNN Pass: change memcpy to std::copy
* Fix MKLDNN FC handling of mismatch input and weights dims
* Lower tolerance for MKL-DNN in resnet50 test
test=develop
* Adjust FC to support MKLDNN Op placement
test=develop
* Adjust Placement Op to set use_mkldnn attribute for graph
test=develop
* MKLDNN FC: fix weights format so that gemm version is called
test=develop
* FC MKLDNN: Remove tolerance decrease from tester_helper
* FC MKL-DNN: Refactor the code, change input reorder to weight reorder
* MKL-DNN FC: Introduce operator caching
test=develop
* FC MKL-DNN: Fix the tensor type in ExpectedKernelType
test=develop
* FC MKL-DNN: fix style changes
test=develop
* FC MKL-DNN: fallback to native on non-supported dim sizes
test=develop
* FC MKLDNN: fix CMake paths
test=develop
* FC MKLDNN: Refine placement pass graph mkldnn attribute
test=develop
* Fix Transpiler error for fuse_conv_eltwise
test=develop
* Fix missing STL includes in files
test=develop
* FC MKL-DNN: Enable new output size computation
Also, refine pass to comply with newest interface.
test=develop
* FC MKL-DNN: enable only when fc_mkldnn_pass is enabled
* FC MKL-DNN: Allow Weights to use oi or io format
* FC MKL-DNN: Adjust UT to work with correct dims
test=develop
* Enable MKL DEBUG for resnet50 analyzer
test=develop
* FC MKL-DNN: Improve Hashing function
test=develop
* FC MKL-DNN: Fix shape for fc weights in transpiler
* FC MKL-DNN: Update input pointer in re-used fc primitive
* Add log for not handling fc fuse for unsupported dims
test=develop
* FC MKL-DNN: Move transpose from pass to Op Kernel
test=develop
* FC MKL-DNN: Disable transpose in unit test
test=develop
* FC MKL-DNN: Remove fc_mkldnn_pass from default list
* Correct Flag for fake data analyzer tests
test=develop
* FC MKL-DNN: Add comment about fc mkldnn pass disablement
test=develop
* FC MKL-DNN: Disable fc in int8 tests
test=develop
* add conv_concat_relu fuse
test=develop
* add test code
test=develop
* added missing include with unordered_map
test=develop
* review fixes for wojtuss
test=develop
* remove 'should (not) be fused' comment statements
one of them was invalid anyway
test=develop
* fix quantize_squash_pass segfault when there is no tensor linked do Bias input
test=develop
* add googlenet test
test=develop
* fix concat CreateKey not using input format
test=develop
* Relu6 is the bottleneck op for Mobilenet-v2. As the mkldnn supports the conv/relu6 fusion, we implement it fusion via cpass way. Due to the int8 enabling for this fusion will be supported in MKLDNN v0.20, so this PR is focused on the fp32 optimization.
Below table shows the benchmark(FPS) which measured on skx-8180(28 cores)
Batch size | with fusion | without fusion
-- | -- | --
1 | 214.7 | 53.4
50 | 1219.727 | 137.280
test=develop
* Fix the format issue
test=develop
* Add the missing nolint comments.
test=develop
* Fix the typos.
test=develop
* Register the conv_brelu_mkldnn_fuse_pass for the MKLDNN engine.
test=develop
* Adjust the indentation.
test=develop
* Add the test_conv_brelu_mkldnn_fuse_pass case.
test=develop
* Slightly update the code per Baidu comments.
Let the parameter definition embedded into the code.
That's will make the code easy to understand.
test=develop
* fix bn fuse vardesc and add model saver
test=develop
* unify save model in test helper
test=develop
* fix mkdir on windows
test=develop
* remove magic number use bn bias var desc
test=develop
* add parallel build script to ci test=develop
* 1. classify the test case as single card/two cards/multiple cards type
2. run test case according to the run type
1. refine anakin engine
2. add data type for zero copy
align dev branch and PaddlePaddle:feature/anakin-engine brach
the cudnn workspace modify was not included for now, because we use a hard code way
in feature/anakin-engine branch. There should be a better way to implement it,
and subsequent submissions will be made.
test=develop
* - Fix to crash of Transformer when mkldnn is to be used
Desc: TensorCopy was not setting MKLDNN primitive descriptor when layout was to be kMKLDNN
test=develop
* - Enable transformer for mkl-dnn
test=develo
* - Compilation fix
test=develop
* - Removed manual selection of MKL-DNN ops to be used in Transformer test
test=develop
* Add cpu_quantize_pass for C-API quantization
test=develop
* add cpu_quantize_pass test
* fix lint: add include memory unorderd_map and unordered_set
test=develop
* fuse_relu 1
test=develop
* tuned 2 without squash
* fixes
test=develop
* remove unused vars
test=develop
* refactored
test=develop
* fix lint c-style cast -> C++ style cast
test=develop
* remove QuantMax and c style casts
test=develop
* last usage of QuantMax removed
test=develop
* Fix Analysis Predictor UT
Check if memory_optimize_pass has already been added
to the analysis config before adding a new one, so
that it is not added multiple times.
test=develop
* change map to unordered_map
fix the forgotten part of cpu_quantize_pass_tester.cc
test=develop
* removed quantized attribute
* fixed cpu_quantize_pass_tester and op attr comments
test=develop
* removed redundant line
test=debug
* removed gmock
test=develop
* fix after merge
* Refine cmake's download function.
test=develop
* Set DOWNLOAD_NO_EXTRACT to 1 pure download function.
test=develop
* Fix unpack problem in ExternalProject_Add, and it seem DOWNLOAD_NO_EXTRACT option is not support in cmake-3.5.
test=develop
* Remove some superfluous std::move calls
The std:move triggered a build error (with -Werror):
```
[ 9%] Building CXX object paddle/fluid/memory/allocation/CMakeFiles/allocator_facade.dir/allocator_facade.cc.o
/home/tej/code/gbuella_paddle/paddle/fluid/memory/allocation/allocator_facade.cc:86:29: error: moving a temporary object prevents copy elision [-Werror,-Wpessimizing-move]
[this] { return std::move(CreateAllocatorWithChunk()); }, capacity);
^
/home/tej/code/gbuella_paddle/paddle/fluid/memory/allocation/allocator_facade.cc:86:29: note: remove std::move call here
[this] { return std::move(CreateAllocatorWithChunk()); }, capacity);
^~~~~~~~~~ ~
1 error generated.
```
See: https://reviews.llvm.org/D7633
* Remove a superfluous lambda capture from framework/operator.h
```
[ 10%] Building CXX object paddle/fluid/platform/CMakeFiles/device_context.dir/init.cc.o
In file included from /home/tej/code/gbuella_paddle/paddle/fluid/platform/init.cc:19:
/home/tej/code/gbuella_paddle/paddle/fluid/framework/operator.h:229:21: error: lambda capture 'this' is not used [-Werror,-Wunused-lambda-capture]
[this](Variable* var) { return var; });
^~~~
1 error generated.
```
Changing it to `return it->second;`, as is in the function below.
* Rethrow an exception (instead of copying it)
```
[ 11%] Building CXX object paddle/fluid/framework/CMakeFiles/operator.dir/operator.cc.o
/home/tej/code/gbuella_paddle/paddle/fluid/framework/operator.cc:191:13: error: local variable 'exception' will be copied despite being thrown by name [-Werror,-Wreturn-std-move]
throw exception;
^~~~~~~~~
/home/tej/code/gbuella_paddle/paddle/fluid/framework/operator.cc:191:13: note: call 'std::move' explicitly to avoid copying
throw exception;
^~~~~~~~~
std::move(exception)
```
See https://reviews.llvm.org/D43322 for an explanation of this diagnostic message.
* Remove an unused variable
```
/home/tej/code/gbuella_paddle/paddle/fluid/framework/operator.cc:884:16: error: private field 'scope_' is not used [-Werror,-Wunused-private-field]
const Scope& scope_;
^
```
* struct ComputationOpHandle -> class ComputationOpHandle
```
[ 13%] Building CXX object paddle/fluid/framework/details/CMakeFiles/memory_early_delete_pass.dir/memory_early_delete_pass.cc.o
In file included from /home/tej/code/gbuella_paddle/paddle/fluid/framework/details/memory_early_delete_pass.cc:21:
/home/tej/code/gbuella_paddle/paddle/fluid/framework/details/reference_count_pass_helper.h:30:1: error: class 'ComputationOpHandle' was previously declared as a struct; this is valid, but may result in linker errors under the Microsoft C++ ABI [-Werror,-Wmismatched-tags]
class ComputationOpHandle;
^
/home/tej/code/gbuella_paddle/paddle/fluid/framework/details/computation_op_handle.h:29:8: note: previous use is here
struct ComputationOpHandle : public OpHandleBase {
^
/home/tej/code/gbuella_paddle/paddle/fluid/framework/details/reference_count_pass_helper.h:30:1: note: did you mean struct here?
class ComputationOpHandle;
^~~~~
struct
1 error generated.
```
* Fix name() methods under fluid/operators
```
In file included from /home/tej/code/gbuella_paddle/paddle/fluid/operators/jit/gen/act.cc:15:
In file included from /home/tej/code/gbuella_paddle/paddle/fluid/operators/jit/gen/act.h:19:
/home/tej/code/gbuella_paddle/paddle/fluid/operators/jit/gen/jitcode.h:71:23: error: 'name' overrides a member function but is not marked 'override' [-Werror,-Winconsistent-missing-override]
virtual const char* name() const = 0;
^
/home/tej/code/gbuella_paddle/paddle/fluid/operators/jit/gen_base.h:31:23: note: overridden virtual function is here
virtual const char* name() const = 0;
^
```
test=develop
test=develop
size_t is an unsigned integer, with a conversion rank
larger than int, therefore in the following expression
the int value was promoted to size_t, making it a
subtraction of unsigned values. The result of such
a subtraction is also an unsigned value.
翟飞跃
Wojciech Uss
lidanqing
石晓伟
Michał Gallus
Zhaolong Xing
mozga-intel
Tao Luo
Jacek Czaja
Sylwester Fraczek
Zhen Wang
Zeng Jinle
Michał Gallus
lijianshe02
guomingz
liuwei1031
wopeizl
flame
Leo Zhao
tensor-tang
tangwei12
wopeizl
nhzlx
gongweibao
root
superjomn
Yihua Xu
bingyanghuang
chengduo
Michał Gallus
flame
Qiyang Min
minqiyang
Liu Yiqun
Dun
Gabor Buella
dzhwinter
qingqing01
Jiabin Yang