* add support of matmul with multiple head even different width and height
Original matmul with multiple head supports only the mat_a.width == mat_b.height,
in that case, mat_b will be horizontally split. In this patch, we extend the
support when mat_a.width != mat_b.height but mat_a.width/head_number == mat_b.height,
in this case, mab_b will be vertically split.
One example is A is [3, 8], B is [2, 16], head_number is 4. In this
case, A will be split as [3, 2], B will be (vertically) split as
[2, 4]. The final result will be 4 matrix of 4 matrix of [3,4], i.e. [3, 16]
test=develop
* add support of matmul with multiple head even different width and height
Original matmul with multiple head supports only the mat_a.width == mat_b.height,
in that case, mat_b will be horizontally split. In this patch, we extend the
support when mat_a.width != mat_b.height but mat_a.width/head_number == mat_b.height,
in this case, mab_b will be vertically split.
One example is A is [3, 8], B is [2, 16], head_number is 4. In this
case, A will be split as [3, 2], B will be (vertically) split as
[2, 4]. The final result will be 4 matrix of 4 matrix of [3,4], i.e. [3, 16]
test=develop
* refactor the code of matmul with multiple head even different width and height
test=develop
* Add support for new QAT models
test=develop
Co-Authored-By: Michał Gallus <michal.gallus@intel.com>
Co-Authored-By: Wojciech Uss <wojciech.uss@intel.com>
* fixed fps results
test=develop
* fix top5 accuracy drop problem
* updated for new QAT models
* skip quantizing average pooling - dirty but working
* add missing pass
* added missing conv+brelu fuse pass
* removed a call to non-existent pass
test=develop
* renamed pass
test=develop
* Adjust finding pooling scale to newest QAT models
* Remove unnecessary code from quantization_mkldnn_pass
* Copy Pooling input scale to output scale in QAT
* Refactor & remove unused code in QAT
* Incorporate fp32 FC into QAT
test=develop
* Enable graph drawing with debug flag
test=develop
* Add tests for QATv2
* Fix paths for QATv2 models
test=develop
* Add option to save transformed int8 qat model
test=develop
* Remove redundant lines from qat mkldnn pass
test=develop
* Delegate disablement of avg pooling to qat
test=develop
* fix CI bug, test=develop
* Follow Wangzhen's Review, test=develop
* Update API.spec
test=develop
* Name False in (is_unsigned, TensorScale) tuple
test=develop
* Remove constraint that last dimension is forced to be 1 by add
lookup_table_v2 test=develop
* modify into PADDLE_ENFORCE_CUDA_SUCCESS test=develop
* Revert "modify into PADDLE_ENFORCE_CUDA_SUCCESS test=develop"
This reverts commit 8a960bfc61e51aa27c3c529df8fb90b93ebd19f9.
* move api into fluid.embedding test=develop
* fix example code test=develop
* move one_hot into fluid.one_hot
* modify api.spec test=develop
* fix loss shape test=develop
1. Support customize eval function instead of eval program.
2. Fix loading checkpoint in quantization strategy.
3. Support saving eval model when saving a checkpoint.
4. Fix decoder of loading context in PaddleSlim.
5. Fix restoring from the checkpoint of uniform prune strategy.
6. Support saving eval model and infer model during training.
7. Add ‘unitest’ for saving eval model, saving infer model and uniform pruning restoring from the checkpoint.
8. Fix pruning of depthwise_conv_grad op by updating the groups.
* support change shuffle thread num
* support change train thread num
* fix receive shuffle data of each channel
* data norm stop gradient
* add check thread_tensor type and root_tensor type when merge metric
* remove sleep in shuffle, add config
* add config of pslib client to client communication
* fix xbox str
* add data norm op testcase
* add flush in trainer finalize
* make OpTest check grad inplace even if forward has no inplace, test=develop
* do not run PE when enable_inplace is False, test=develop
* add conv3d cuda kernel for float16 type, test=develop
* refactor OpTest for inplace, test=develop
* add comments, test=develop
* add recompute based checkpoints methods for large batch training
test=develop
* add append_backward_with_forward_recomputation
test=develop
* refine optimizer
test=develop
* update backward and optimizer
test=develop
* make Variable usable
test=develop
* add recompute code
* refine optimizer
test=develop
* refine addup _append_backward_ops_with_checkpoints_
1) for recompute part, just cache the grad_op_desc without appending to block
2) before appending grad_op_desc to backward part, addup_repetitive_vars, remove unused branch
test=develop
* make method private
* add recompute strategy into DistributedStrategy
test=develop
* checkpoint version3
test=develop
* remove some print information
test=develop
* remove unused sumop
test=develop
* try to fix recompute with graph building modules
* add input names to vars should be held
* add memory debug tool
* backup backward
* Fix bugs
* add backward desc for op not in any segments
* add exception info for sub_block
test=develop
* modify code style
test=develop
* modify code style
test=develop
* remove print functions
test=develop
* add API spec
test=develop
test=document_preview
* make Recompute a child class of Optimizer
test=develop
test=document_preview
* add API spec
test=develop
test=document_preview
* modify API spec
test=develop
test=document_preview
* add document for Recompute
test=develop
test=document_preview
* change API doc of Rcompute
test=develop
test=document_preview
* code cleaning
test=develop
test=document_preview
* modify API spec
* fix bugs when segments hold no element
* add testcase for Recompute Optimizer
test=develop
test=document_preview
* add test for apply_gradient, and code cleaning
test=develop
test=document_preview
* add test case for load function
* enable CI
test=develop
test=document
* add test case
test=develop
test=document_preview
* add sample code for 4 function of recompute optimizer
test=develop
test=document_preview
* move tree_conv to fluid.contrib.layers
test=develop
* update API.spec for tree_conv
test=develop
* update tree_conv api to increase unit coverage
test=develop
* refactor dygraph,test=develop
* fix failed unittest,test=develop
* polish code,test=develop
* check windows ci error,test=develop
try to fix windows ci error by np.allclose,test=develop
* polish vlog and profiler, test=develop
* try to fix preceding ops order,test=develop
* test transformer in windows ci, test=develop
* use python c-api to speed up tracer.trace,test=develop
* test=develop, fix docker with paddle nccl problem
* test=develop, add ut for debug string and gradient_accumulator
* test=develop, add tests for layer/gradient_accumulator/prepared_op
* test=develop, fix complie error for test_prepared_op
* test=develop, add more ut for dygraph
* test=develop, create API.spec for dygraph api change
* test=develop, refoctor name to make it easier to understand
* test=develop, refoctor name to make it easier to understand
* test=develop, fix multi-gpu failed problem , add Tracer tests, change PADDLEENFORCE to PADDLEENFORCE_EQ
* test=develop, fix ut failed on parallel se-resnext
* test=develop, change one more PADDLE_ENFORCE
* support auto prune in dygraph mode
* test=develop, support auto prune
* test=develop, merge develop conflict
* test=develop, fix test_layer and test_tracer ut
* test=develop, fix bug which may cause stop_gradient disabled with a list of backward inputs
modified interpolate_op to support tensor attribute
1. the parameter out_shape of image_resize、resize_nearest/bilinear/trilinear can be a list or a 1-D tensor variable. If a list, each element can be an integer or a tensor variable with shape: [1].
2. the parameter scale of above Ops can be a 1-D tensor variable.
modified document of image_resize, resize_nearest, resize_bilinear, resize_trilinear and add some code example.
add crop_tensor op. The main difference with crop is :
1. If the argument shape is a list, each element is an integer or a tensor variable with shape: [1]. This way is suitable for the case that the shape may be changed each iteration.
2. If the argument shape is a variable. Its rank must be 1. In crop op, the rank of shape must be the same as x
offsets can be a list, in which each element is an integer or a tensor variavle with shape: [1].
* Add fc_elementwise_layernorm_fuse pass and unittest.
* Add fused_fc_elementwise_layernorm op and its GPU kernel.
test=develop
* Apply fc_elementwise_layernorm_fuse_pass to GPU inference.
* Add the setting of attrs in the definition of binary_op.
test=develop
* Add comment.
* Implement the unittest.
test=develop
* Change the unittest name of layer_norm.
test=develop
* strided_slice op basic function test=develop
* test=develop rewrite and fix
* fix bug test=develop
* fix for the PADDLE_ENFORCE usage
* add some unit testw
* fix for the aip test and copright and fix test=develop
* fix API.spec test=develop
* fix API.spec test=develop
* add axis parameter test=develop
* fix for the build error test=develop
* fix python api test=develop
* fix the build test=develop
* fix build test=develop
* fix API spec test=develop
* test=develop add some comment and single op test
* fix API spece test=develop
* fix test=develop
* fix test=develop
* fix api test=develop
* fix api test=develop
* fix API.spec test=develop
* fix typo test=develop
* fix API.spec test=develop
* fix API typo test=develop
* fix doc and API.spec test=develop
improve pow op according to reviews:
1. Delete unnecessary judgement statements in PowGradOpDescMaker;
2. Improve test of test_api;
overload GetKernelTypeForVar
add stop_gradient=True when attr(factor) is tensor Variable, change examples in API pow.
test=develop,test=document_preview
add support parameter inference when argument shape is a list containing integer and tensor variable;
test=develop
fix reshape op according to reviews:
1. improve or message;
2. improve test of test_api.
test=develop,test=document_preview
fix reshape op: Add error message in nn.py, test=develop
add stop_gradient=True when attr(shape) is tensor Variable.
change examples in API reshape.
test=develop,test=document_preview
add support parameter inference when arguments starts or ends is a list containing integer and tensor variable;
test=develop,test=document_preview
improve slice op according to review(from hongyu). test=develop
fix slice op according to review: infer_flags, test=develop
fix slice op: improve overload operator __getitem__ to support attrs(starts and ends) are Variable.
test=develop,test=document_preview
fix test_slice_op: add TestSliceOp_decs_dim_6 to resolve conflict with test_slice_ngraph_op. test=develop
add stop_gradient=True when attr(starts) or attr(ends) is tensor Variable.
test=develop,test=document_preview
1. add tensor support for argument expand_times in expand op;
2. add support parameter inference when argument expand_times is a list containing integer and tensor variable;
improve expand op according to reviews:
1. add doc of ExpandTimes in expand_op.cc;
2. improve the test of test_api.
add stop_gradient=True when attr(expand_times) is tensor Variable, change code examples.
test=develop,test=document_preview
* refactor dygraph,test=develop
* fix failed unittest,test=develop
* polish code,test=develop
* check windows ci error,test=develop
try to fix windows ci error by np.allclose,test=develop
* polish vlog and profiler, test=develop
* try to fix preceding ops order,test=develop
* test transformer in windows ci, test=develop
* use python c-api to speed up tracer.trace,test=develop
* test=develop, fix docker with paddle nccl problem
* test=develop, add ut for debug string and gradient_accumulator
* test=develop, add tests for layer/gradient_accumulator/prepared_op
* test=develop, fix complie error for test_prepared_op
* test=develop, add more ut for dygraph
* test=develop, create API.spec for dygraph api change
* add transform_data to dygraph
* test=develop, refoctor name to make it easier to understand
* test=develop, refoctor name to make it easier to understand
* add test and change input to const ref for safety
* test=develop, fix multi-gpu failed problem , add Tracer tests, change PADDLEENFORCE to PADDLEENFORCE_EQ
* add ut for data transform
* refine ut for data_transform
* test=develop, fix ut failed on parallel se-resnext
* test=develop, change one more PADDLE_ENFORCE
* add test_tracer on multiple devices
* test=develop, change place to mutable for data transform
* test=develop, add transform data on same place test and remove useless log
* test=develop, Add to do for data layout and and ut for conv2d with no bias
* Implement the operator with sprase matrix multiply
* Update the URL of mklml library.
test=develop
* Disable MKLML implematation when using no-linux.
test=develop
* optimize bp with mkl sparse matrix
test=develop
* tmp add fused_emb_seq layer
* Add the support of padding_idx attribute.
test=develop
* add padding_idx support
test=develop
* implement grad refer lego
test=develop
TemporaryAllocator is a singleton used for allocating memory for Cudnn. Since it is a singleton, we can delete it for better performance in memory.
We replace TemporaryAllocator by CUDADeviceContextAllocator and CUDADeviceContextAllocation, which uses stream callback to delete the memory allocated for the stream to avoid singleton.
Also added data_feed_proto to operator to fix CI in CPU compilation
* Refine the codes related to fc op.
* Add GPU implementation for fc functor.
* Apply fc_fuse_pass in GPU inference.
test=develop
* Change the cmake for fc op.
* Change PADDLE_ENFORCE to PADDLE_ENFORCE_EQ.
* Add an attribute to set the activation type in fc_op.
* Enhance the unittest of fc_op.
test=develop
* Remove the declaration of FCOpGrad back to the header file.
test=develop
* Set default value for newly added arguments in test_fc_op.
test=develop
* Remove constraint that last dimension is forced to be 1 in huber_loss
test=develop
* add y[rank-1] == 1 when x_rank=y_rank test=develop
* modify into contain_unknown_dim test=develop
* refactor dygraph,test=develop
* fix failed unittest,test=develop
* polish code,test=develop
* check windows ci error,test=develop
try to fix windows ci error by np.allclose,test=develop
* polish vlog and profiler, test=develop
* try to fix preceding ops order,test=develop
* test transformer in windows ci, test=develop
* use python c-api to speed up tracer.trace,test=develop
* test=develop, fix docker with paddle nccl problem
* test=develop, add ut for debug string and gradient_accumulator
* test=develop, add tests for layer/gradient_accumulator/prepared_op
* test=develop, fix complie error for test_prepared_op
* test=develop, add more ut for dygraph
* test=develop, create API.spec for dygraph api change
* test=develop, refoctor name to make it easier to understand
* test=develop, refoctor name to make it easier to understand
* test=develop, fix multi-gpu failed problem , add Tracer tests, change PADDLEENFORCE to PADDLEENFORCE_EQ
* test=develop, fix ut failed on parallel se-resnext
* test=develop, change one more PADDLE_ENFORCE
Bob Zhu
Liufang Sang
Tao Luo
Wojciech Uss
Aurelius84
Yi Liu
jhjiangcs
Yang Zhang
Aurelius84
whs
xujiaqi01
Kaipeng Deng
Ghost Under Moon
Zeng Jinle
Leo Chen
juncaipeng
Zhang Ting
mapingshuo
chengduo
wopeizl
ruri
tangwei12
lvmengsi
Adam
Jiabin Yang
flame
gongweibao
Yiqun Liu
WangXi
Jie Fang
wangchaochaohu
Huihuang Zheng
123malin
Bai Yifan
LielinJiang
chengjuntao
liym27
翟飞跃
zhongpu
Chen Weihang
JesseyXujin
Huihuang Zheng
Youwei Song
Thunderbrook
wangguanzhong