99d30bfc36
speedup slice impl ( #23340 )
...
test=develop
5 years ago
1a6ce8b910
add swish split gelu plugin dynamic support ( #23305 )
...
test=develop
5 years ago
2bb1b0e89e
[DNNL] Added MKL-DNN inplace pass for C-API inference ( #23315 )
5 years ago
0471476a18
fix nccl comm double free bug ( #23344 )
...
As nccl comm is not created by CUDADeviceContext, it should be destroyed by the creator as the best practice of RAII.
5 years ago
1ee2a9a424
Profiler refine ( #23294 )
...
* refine output of profiler for child event
5 years ago
488b2387e2
Feature/expand params in auto-generated pybind functions for dygraph operators ( #23181 )
...
* expand parameters, test=develop
* support resnet, test=develop
* fix resnet, test=develop
* support duplicable out, test=develop
* support ptb
* fix bugs, test=develop
* support null input, test=develop
* fix bugs, test=develop
* fix batchNorm is_test, test=develop
* refine code, test=develop
* follow comments, test=develop
* follow comments, test=develop
* follow comments, test=develop
* follow comments, test=develop
5 years ago
20eed5401a
Change fluid.layers.where‘s C++ operator name ( #23250 )
5 years ago
2169e6fb58
Initialize global nccl_comm in PE ( #23275 )
5 years ago
012886df79
[DNNL] Softmax mkldnn op inplace support ( #23197 )
5 years ago
75ebb48a91
supports thread-binding stream, test=develop ( #23177 )
5 years ago
708ded584e
pause the io_utils_test of int64 and resume after repair, test=develop ( #23234 )
5 years ago
babda94c8a
Distinguish public/private global vars ( #23269 )
...
* distinguish public/private vars, test=develop
* fix windows issues, test=develop
5 years ago
58615a6272
Improve elementwise performance. ( #23001 )
...
* Improve elementwise performance.
Elementwise performace is poor as walk into CommonGradBroadcastCUDA, add some new kernels for different data pattern.
* Add some cuda kernel to speedup common broadcast cases. test=develop
* Add more test cases and fix cuda kernel bug. test=develop
* Remove tests as cpu percision fails.test=develop
* Refine SplitDims, test=develop
* Change file mode, test=develop
5 years ago
f836c8aa8f
add check for scales and a message ( #23119 )
5 years ago
8bfd62ffb7
Expose dygraph.grad api ( #23124 )
...
* expose dygraph.grad api, test=develop, test=document_fix
* add more parameter in dygraph.grad API, test=develop
* add only_inputs=True parameter, test=develop
* follow comments, test=develop, test=document_fix
* fix typo, test=develop, test=document_fix
5 years ago
0129f4b568
Add some inference API comments for AnalysisPredictor ( #23242 )
...
* add inference api doc. test=develop
5 years ago
c00d427d52
simplify the cmake log of ir/CMakeLists.txt ( #23262 )
...
test=develop
5 years ago
77b4dc80c9
code polish for adding const qualifier, test=develop, test=document_fix ( #23248 )
5 years ago
430b0099c9
[Paddle-TRT]: Ernie Dynamic shape support. ( #23138 )
...
* add dynamic plugin support.
test=develop
* change emb eltwise layernorm to math function
test=develop
* add emb eltwise layernorm
test=develop
* can run dynamic shape ernie
test=develop
* fix ci
test=develop
* add ut for trt ernie dynamic
test=develop
* refine dynamic shape c++ interface.
test=develop
* fix comments
test=develop
* fix comments
test=develop
5 years ago
68ea1ad55b
add clear one table ( #23089 )
...
* add clear_one_table
* test=develop
5 years ago
ae3bb16d06
add MaskAucCalculator in paddlebox ( #23157 )
...
* add maskauc in paddlebox; test=develop
5 years ago
6af480ca33
Support int64 for op assign_value. test=develop ( #23179 )
5 years ago
53e6f8e1da
rename macro, test=develop ( #23161 )
5 years ago
bba740710d
add cuda resource pool for BufferedReader, test=develop ( #23152 )
5 years ago
7d8d50b6cc
rename no_need_buffer_vars macro, test=develop ( #23160 )
5 years ago
Liufang Sang
a486a739e1
fix compile error in win gpu ( #23196 )
...
* fix compile error in win gpu test=develop
* fix compile error in win gpu test=develop
* fix compile error in win gpu test=develop
5 years ago
7ca77a90ac
add Tensor::IsSharedBufferWith method, test=develop ( #23175 )
5 years ago
b8886bf122
rename no_need_buffer_vars_macro, test=develop ( #23159 )
5 years ago
bae5930ba1
fix graph attr copy issues, test=develop ( #23191 )
5 years ago
b721e23b25
transpose cudnn using cudnn v7 api ( #19738 )
...
* refine the transopose conv using v7 to choose algorithm
5 years ago
46b8d282dc
Add some inference API comments for AnalysisConfig ( #23117 )
...
* add some API comments in paddle_analysis_config.h, test=develop
* add some API comments in paddle_analysis_config.h, test=develop
5 years ago
4f5e4540f8
Improve SGD jit code to work with large data ( #23120 )
5 years ago
Liufang Sang
4db031902d
add dequantize_log_op and make pyramid hash support int8 weight ( #22548 )
...
* add dequantize_log_op and make pyramid hash support int8 weight test=develop
* add unittest and update pyramid hash op test=develop
* remove paddle_enforce test=develop
* fix error message test=develop
* remove incorrent commit test=develop
* fix error message in log_dequantize test=develop
* change 2019 to 2020 test=develop
* remove useless check_grad test=develop
5 years ago
e5fef8f38a
[Dygraph double grad]Code polish ( #23121 )
...
* fix dygraph double grad, test=develop
* fix unpack constructor, test=develop
5 years ago
9258e96094
fix read op comments, test=develop, test=document_fix ( #23122 )
5 years ago
acfc9b8a70
Reader sequential and inference partial feed ( #22699 )
...
* sequential reader stage 1, test=develop
* fix ut, test=develop
* fix iterable=False reset bug, add some logs and polish code, test=develop
* inference feed partial data, test=develop
* Turn on keep_order=True for test, test=develop
* enhance ut to test more cases, test=develop
* test commit for reverting
* Revert "test commit for reverting", test=develop
This reverts commit 80aef42ef52ba1ee79627d6f663a624ec4f12f58.
* add ut of merged and unmerged results, test=develop
* add more uts for coverages and add en doc of api, test=develop
* follow comments, test=develop
* change note style, test=develop
5 years ago
95b356a069
update embedding_eltwise_layernorm fuse and kernel. test=develop ( #23114 )
...
update embedding_eltwise_layernorm fuse pass and fused kernel, to support multi input
5 years ago
a31d7328b7
Add dygraph double grad implementation ( #22939 )
...
* add double grad implementation for dygraph, test=develop
* polish code, add uts, test=develop
* fix place bug, test=develop
* polish codes, add more uts for coverages, test=develop
* add no_grad_set, test=develop
* add star gan ut, test=develop
* follow comments, test=develop
5 years ago
3af4771122
Add the detection and code-generation of sqrt and square in fusion_group ( #23095 )
5 years ago
0c30098f8b
Add need_save_delta parameter to solve OOM ( #23097 )
5 years ago
2e2da7124b
high-performance dygraph slice ( #22879 )
...
* move __getitem__ to cpp
* bug fix
* add type check and gil release
* support negative step with omitted ends
test=develop
* code refine
test=develop
* bug fix
test=develop
* slice always return different pyobj
test=develop
5 years ago
abee05a8c8
added mkldnn swish activation ( #23041 )
5 years ago
8c6fde9e69
fix align error ( #23090 )
...
test=develop
5 years ago
Liufang Sang
915b892a15
Fix div zero in fake quantize op ( #22966 )
...
* fix div zero test=develop
* fix div zero test=develop
* add hostdevice function test=develop
* add eps when is zero test=develop
5 years ago
121b2aed4d
initialize global nccl context in dygraph ( #23037 )
...
initialize global nccl context in dygraph
test=develop
5 years ago
880eb04d93
skip PrepareData when it is unnecessary ( #22839 )
...
* remove unnecessary prepare data, test=develop
* Op in while block will not skip PrepareData, test=develop
5 years ago
01ab8a0619
add approximation for gelu, test=develop ( #22961 )
...
add approximation for gelu, default value is False (only kernel with eigen is added, remove code for computing gelu with MKLDNN temporarily)
5 years ago
5842ae6785
Revert "Change ShareDataWith() to TensorCopy() in conv_mkldnn ( #22695 )" ( #22985 )
5 years ago
24db750386
fix trt int8 calib precision bug. test=develop ( #23036 )
5 years ago
1dc1f9270e
Fix lod error of concat op for axis = 0 ( #22538 )
5 years ago
660ff18488
fix datsset test=develop ( #23043 )
5 years ago
714b0076b6
Override GetKernelTypeForVar to avoid device transform, test=develop ( #23032 )
5 years ago
112e3edbf6
fix the conv group problem test=develop ( #23025 )
5 years ago
db40ee86db
fix unittets. test=develop ( #23018 )
5 years ago
99db0cf762
remove debug log test=develop ( #22994 )
5 years ago
3757e0687c
Add Unittest for backward of fusion group ( #22932 )
...
* add fusion group test for backward and refine code
5 years ago
63f3ada7b9
fix bug which input shape ( #22965 )
...
* fix bug which input shape, test=develop
* add error type,test=develop
5 years ago
137d6563fc
add check for assigned data, test=develop ( #22960 )
5 years ago
f0d193a23c
Cast fusion for fusion group ( #22876 )
...
* add support for expression type convert and add cast Op support in fusion group
5 years ago
29a7a52d38
Fix instag ( #22632 )
...
* update
* update test=develop
* update compile set test=develop
* update compile set test=develop
* update test=develop
* update test=develop
* update test=develop
* update compile setting test=develop
* update compile setting test=develop
* update run demo test=develop
* update test=develop
* update test=develop
* fix test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update format test=develop
* update format test=develop
* update style test=develop
* update style test=develop
* change style test=develop
* change style test=develop
* change style test=develop
* add dataset unittest test=develop
* update test=develop
* update for record test=develop
* udpate style for record test=develop
* update for record test=develop
* update for record test=develop
* update for record test=develop
* fix format test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* fix compile warning test=develop
* add attr default test=develop
* add unittest test=develop
* fix style test=develop
* fix style test=develop
* change out_val_ifempty to out_val_if_empty test=develop
5 years ago
c979c9f2b0
refine the profiler print test=develop ( #22968 )
5 years ago
ff3ddbb502
add skip_layernorm pass. test=develop ( #22895 )
...
* add skip_layernorm pass. test=develop
5 years ago
f154d5860f
Speed up the matmul op, use the gemm replace the batch gemm ( #22926 )
...
In the op of gemm, we use the gemm to replace batch gemm, speed up the matmul op
5 years ago
056edf3929
Change ShareDataWith() to TensorCopy() in conv_mkldnn ( #22695 )
5 years ago
8d6dc102fe
[Ernie GPU Optimize]: Embedding_eltwise_layernorm Fuse ( #22494 )
...
* 1. add embedding eltwise layernorm fuse
2. add embedding eltwise layernorm op
3. refine inplace_add_relu
4. refine fc_eltwise_layernorm
test=develop
* 1. refine fc
test=develop
* fix comments
test=develop
* fix comments
test=develop
5 years ago
3d8571e884
modify assign op and add unittest of assign op ( #22769 )
...
As the title.
5 years ago
d33c4343e1
Imperative tracer refactoring ( #22457 )
...
* refine grad maker, test=develop
* refactor tracer stage 1, test=develop
* merge develop to solve conflict third times, test=develop
5 years ago
61fef9754b
Fix fc padding bug during inference fusion ( #22860 )
...
* fix fc padding during fusion, test=develop
* fix optim model inference after SaveOptimModel, test=develop
5 years ago
ad9c8f6d2d
fix communicator when break under pyreder mode ( #22911 )
...
* fix communicator when breaking under PyReader mode, test=develop
* revert some vlog level to 0, test=develop
5 years ago
5ba9dfc16a
add lookup_table_dequant_op ( #22900 )
...
add lookup_table_dequant_op
5 years ago
a020a25797
Fix model int8 quant fail, test=develop ( #22891 )
...
As model fails when enable int8 quant, so disable allocate memory in cpu
for small variable.
5 years ago
dd67d44a50
[Paddle-TRT] : (Part1) Dynamic shape support ( #22868 )
...
* change the ci trt from version 5. to 6.0
* paddle-trt dynamic shape support init
* conv+bias or conv+bn dynamic shape support
test=develop
* modity trt engine opconvert
test=develop
* fix ci error
test=develop
5 years ago
07e13b84cd
remove vlog, test=develop ( #22898 )
5 years ago
ca9c8b417d
fix compute ratio of profile, test=develop ( #22872 )
5 years ago
dbb0b9b3b6
refine the profiler print ( #22823 )
...
* refine the profiler print test=develop
5 years ago
0038bfbd1d
Prevent loading of warmup data in analyzer_int8 if enable_int8 is set to false ( #22857 )
5 years ago
1644926a6c
Polish detail implement of dygraph data loader ( #22878 )
...
* polish detail implement of data loader, test=develop
* solve coverage ci problem, test=develop
5 years ago
f686310d81
fix concat_mkldnn op. test=develop ( #22692 )
...
fix concat_mkldnn op when encounter extreame conditions.
5 years ago
5191e54494
reduce default attrs for dynamic graph ( #22850 )
...
* reduce default attrs for dynamic graph, test=develop
* add some explanations for explicit attr, test=develop
* tweak explicit attr comments, test=develop
5 years ago
1a533ed2de
[BUG]: Multihead matmul op's ouput size should be BxSx(N*H) ( #22848 )
...
test=develop
5 years ago
c736fef93b
dygraph backward engine accelerate ( #22808 )
...
* fix loaded program load bug; test=develop
* first version
* speed backward engin; test=develop
* remove useless code; test=develop
* reconvery io.py; test=develop
* remove useless code; test=develop
* remove useless code; test=develop
5 years ago
d41d802ba3
Add flags to limit gpu memory ( #22793 )
...
* add recorded cuda memory apis, fix typo, test=develop
* add more ut, test=develop
* follow comments, test=develop
* fix py35 incompatible issues, test=develop
5 years ago
1861ca88f1
serialize the PaddleTensor, test=develop ( #22810 )
...
* encapsulate the PaddleTensorToLoDTensor, test=develop
* serialize the pd_tensor, test=develop
* serialize tensors to file, test=develop
5 years ago
72ff5a09c3
fix print bug of profile, test=develop ( #22804 )
5 years ago
4e8bc02461
add fluid.device_guard to specify the device type for Op ( #22254 )
...
* add fluid.device_guard to specify the device type for Op
5 years ago
ddb9b46fec
change the function in op_teller, test=develop ( #22794 )
...
* change the function in op_teller, test=develop
* correct the commit-id, test=develop
5 years ago
89cfa49156
Unmerged fetch list ( #22635 )
...
* update ScopeBufferedSSAGraphExecutor&AsyncSSAGraphExecutor&ThreadedSSAGraphExecutor&FastThreadedSSAGraphExecutor&ParallelSSAGraphExecutor&ParallelExecutor for fetching unmerged results.
* add the unit test for fetch_unmerged.
* update ut for multi-card and multi-cpu.
* add the error message and the user suggestion in FetchOpHandle. test=develop
5 years ago
8456c3f4dd
polish the profiler_help code ( #22811 )
5 years ago
2fd1ec1e3e
fix docker build for paddle openblas, test=develop ( #22795 )
5 years ago
7d8d573453
Speed up dygraph DataLoader based on shared memory and LoDTensor serialization ( #22541 )
...
* add lodtensor share memory & serialization, test=develop
* fix windows compile error, test=develop
* deal vartype pickle & fix unittest matching error message, test=develop
* update timeout variable name, test=develop
* refactor memory map implement, test=develop
* clear mmap file discripter when exit unexpectedly, test=develop
* remove the child process fd in advance, test=develop
* remove mmap fds after Queue.put in child process, test=develop
* add hard unittests for register exit func, test=develop
* fix python2 compatibility problem in unittest, test=develop
* fix exception unittest error, test=develop
* polish code based review comment, test=develop
5 years ago
324f2b3922
Fix inference c api PD_GetZeroCopyOutput lod ( #22768 )
...
* fix inference c api lod, test=develop
* fix capi lod problem and enrich tests, test=develop
* delete useless header files and alter const_cast, test=develop
5 years ago
7578fcbac4
Profile code refine ( #22800 )
...
* add profiler_help.h to refine the code test=develop
5 years ago
53a2b68f4e
support customized download command in dataset ( #22782 )
...
* user can call dataset.set_download_cmd to set its customized download cmd
* add UT to cover this scenario
5 years ago
ca9e77a8d4
add sum op support for fusion group ( #22771 )
...
* Add the codegen and auto fusion for sum Op in fusion group
5 years ago
433cef03e5
fix typo word ( #22784 )
5 years ago
ebc7ffc300
fix detection_map. test=develop ( #22705 )
5 years ago
72dde4abde
Refine adam op to improve performance, test=develop ( #22346 )
...
* Refine adam op, test=develop
* Fuse kernels together to reduce cpu time.
* Refine paddle enforce, test=develop
* Remove some comments, test=develop
* Refine code,test=develop
* Refine cuda kernel, test=develop
* Refine code according to comments, test=develop
5 years ago
f2d1cd119a
fix lod level, test=develop ( #22755 )
5 years ago
79d712346f
Correct CPU gradients of the argsort op ( #22739 )
...
* Correct CPU gradients of the argsort op, form a network to test its forward and backward process, test=develop
* fix dynamic threshold error in test_argsort_op, test=develop
5 years ago
2b80e9a719
Add cpu_info without XBYAK ( #22716 )
5 years ago
ae8b5f11a3
Change ShareDataWith() to TensorCopy() in ref_by_trainer_id ( #22717 )
...
As the title
5 years ago
71ab0458e1
Fix pointer and c-api encapsulation ( #22663 )
...
* refine pointer and c-api prototype, test=develop
* fix new c api profile bug, test=develop
* add unit tests, test=develop
5 years ago
b2c1be851a
support cond in clone, test=develop ( #22657 )
...
* support cond in clone, test=develop
* refine code, test=develop
* refine code, test=develop
* follow comments, test=develop
* refine code, test=develop
5 years ago
f97f3f9301
add framework overhead ratio in profile report ( #22590 )
...
* add framework overhead ratio, test=develop
* print GpuMemcpy overhead, test=develop
5 years ago
160d0f1308
fix the CI risk that network cannot be connected ( #22736 )
5 years ago
15c2667143
register fp16 for assign op ( #22744 )
...
* register fp16 for assign op, test=develop
* add op test for fp16, test=develop
5 years ago
882e7f7c3b
Directly getting API.spec for tools/sampcd_processor.py ( #22728 )
5 years ago
1c0653462d
fix generate_mask_labels lod level ( #22743 )
5 years ago
ba140222d6
fix compile&runtime lod_equality of lod_reset ( #22737 )
5 years ago
175954d894
PaddleBox Framework Part2 ( #22466 )
...
* Add two types of Metric Calculator: MultiTaskCalculator & CmatchRankCalculator.
* Add a config for DynamicAdjustChannelNum function to denote whether we will discard the remaining instances when they are not be distributed evenly.
* Remove CPU code in Pull/PushSparse and we will add it back when testing it fully.
* Fix some known issues: such as copying persistable vars after one epoch running.
5 years ago
3132681e8a
add partial_sum op in contrib ( #22292 )
...
* add partial_sum_op, test=develop
* modify the Paddle Error Message, test=develop
* modify the Paddle Error Message, test=develop
* modify the bug for python3, test=develop
* modify the ut for ci, test=develop
* mv to contrib, test=develop
* use check_variable_and_dtype, test=develop
* fix ci, test=develop
* fix conflict, test=dvelop
* add partial concat, test=develop
* fix the conflict, test=develop
* fix the error, test=develop
* rm SSE4, test=develop
5 years ago
611411b90e
Fusion group profile support ( #22718 )
...
* add support for the driver api callback and fix the profiler name show bug
5 years ago
e136661304
add partial_concat op in contrib ( #22528 )
...
* add partial_concat, test=develop
* fix the grids and blocks, test=develop
* fix the Paddle_Enforce, test=develop
* fix the doc of op, test=develop
* fix the doc, test=develop
* fix the doc of the op, test=develop
* replace -1 with None, test=develop
5 years ago
cdf5f6fb8c
Add an inference interface to disable FC padding ( #22097 )
...
* Add an interface of disabling FC padding
* fix bert regression
* polish fc padding interface
* recover pass function
* fix argument error
* fix mkldnn error
5 years ago
d2ba91aad1
fix typo words ( #22653 )
5 years ago
6e7bfe30a6
register fp16 kernel for some ops ( #22650 ) ( #22696 )
...
test=develop
5 years ago
66a3150135
SYNC with communicaotor ( #22344 )
...
* add sync communicator and implement
5 years ago
22bbd54719
Add the support of fp16 in fusion_group ( #22239 )
5 years ago
d97475d53b
fix CPU C inference API compile bug ( #22702 )
5 years ago
adfa5b8354
Add PADDLE_ENFORCE to Check Sequence Length of RecurrentOp ( #22673 )
...
1. Add PADDLE_ENFORCE to Check Sequence Length of RecurrentOp.
2. Also enrich PADDLE_ENFORCE error messages.
5 years ago
74eb82de19
fix go api bug ( #22669 )
5 years ago
a089072c8b
fix the profile print error ( #22665 )
...
* fix the profile print error test=develop
5 years ago
d926214535
[UT coverage] improve the mul_mkldnn_op line coverage ( #22408 )
...
* improve the mul_mkldnn_op line coverage
test=develop
* remove fp32 mul mkldnn kernel
test=develop
* locally refactoring
test=develop
* change according to reviews
test=develop
5 years ago
c65c6ae534
add flag to control profile level in python API ( #22319 )
...
* add python flag to control profile level test=develop
5 years ago
00594c1c88
support dumping params/grads in transpiler mode ( #22490 )
5 years ago
a06d75a280
[Paddle-TRT] Refine the error log about runtime batch and max_batch_size. ( #22535 )
...
* fix trt log
test=develop
* fix comments
test=develop
5 years ago
608447bfd5
Update MKLDNN to v1.2 ( #22521 )
5 years ago
ab610a34ff
transpose_mkldnn code change to meet Paddle standards ( #22591 )
5 years ago
8f035fb637
Add TopK Op Grad CPU&GPU Kernel test=develop ( #22628 )
...
* Add TopK Op Grad CPU&GPU Kernel test=develop
* Add TopK Op Grad, modify grad op maker test=develop
* Add TopK Op Grad, modify grad op maker test=develop
* Add TopK Op Grad, modify PADDLE_ENFORCE test=develop
* Add TopK Op Grad, modify PADDLE_THROW test=develop
* Add TopK Op Grad, modify unittest test=develop
* fix ngraph top k op unittest test=develop
5 years ago
90ee366653
update ops's unittest data type from float32 to float64 and shape over 100 ( #22544 )
...
* update ops's unittest of elementwise_pow, elementwise_max, elementwise_min, scale and sqrt
1. update elementwise_pow, elementwise_max and scale's unitests with input data type (float32 -> float64)
2. fix bug that the elementwise_pow doesn't meet threshold requirements with tackling float64 data
3. remove sqrt from op_accuracy_white_list.py
4. update the unittests of elementwise_pow, elementwise_max and elementwise_min ops that their input data shape over 100
5. test=develop
* modify the writing style according suggestions
test=develop
5 years ago
f7eafca828
remove python inference warning ( #22602 )
5 years ago
fe685cc185
fix enforce test error, test=develop ( #22610 )
5 years ago
9a8203aa25
fix fc_lstm_fuse when multi sub-graph use same fc_bias. test=develop ( #22551 )
...
当一个模型中有多个fc_lstm子图的时候，且其中fc共用了同一个persistable的bias，此时不应该将bias节点删除，只将非persistable的节点去除即可。
5 years ago
266106da75
Fix mismatch with plus sign in the line ( #22588 )
...
* reproduce match error, test=develop, test=document_fix
* fix mismatch error, test=develop, test=document_fix
5 years ago
1d503e6a9e
Golang inference API ( #22503 )
...
* support golang inference
5 years ago
8acd745c25
[Ernie GPU Optim]: Fuse three fc to multihtead matmul ( #22486 )
...
* 1. optim multihead matmul: fuse three fc to multihtead matmul
test=develop
* fix conflict
test=develop
* fix comments
test=develop
5 years ago
96770f519e
Disable fusion_group for windows and mac in build_strategy. ( #22549 )
...
test=develop
5 years ago
08033c8634
fix traced layer with non persistable vars, test=develop ( #22552 )
5 years ago
31b5464632
Add support for dynamic_decode(while) training. ( #22231 )
...
* Add support for dynamic_decode(while) training. test=develop
* Fix assign_op and tensor_array_read_write_op after solving conflict. test=develop
* Fix test_rnn_decode_api.py. test=develop
* Refine docs for apis in rnn.py. test=develop
* Adjust outputs of dynamic_decode. test=develop
* Remove the force_cpu update in assign_op. test=develop
* Remove the force_cpu update in assign_op. test=develop
* Make RNNCell.get_initial_states support batch_dim_idx argument. test=develop
* Rename _create_array_outof_while as _create_array_out_of_while in rnn.py.
test=develop
5 years ago
b0675c8193
fix bug with compiledProgram ( #22495 )
...
* add thread barrier for the compiled program
5 years ago
4cddb43c5c
Add support for Ernie NLP model to the Slim QAT ( #22506 )
...
* a test for Ernie QAT INT8 accuracy check
test=develop
* Remove NLP comparison test to split PRs
test=develop
* Fix typo and tabs, delete commented lines
test=develop
* re-combine the 2 PRs, test=develop
Co-authored-by: Michał Gallus <sand3r@interia.eu>
Co-authored-by: bingyanghuang <33643817+bingyanghuang@users.noreply.github.com>
5 years ago
58d99247f4
support slice double grad, test=develop ( #22166 )
...
* support slice double grad, test=develop
* merge two doublegradopmaker to one doublegradopmaker,test=develop
* change the shape of slice_OP's unittest, test=develop
5 years ago
1a7962be97
Paddlebox about box_wrapper ( #22497 )
...
Refine PaddleBox Framework, Main functions:
* Add MetricMsg util class, which can calculate metrics like AUC, bucket_error, COPC.
* Replace FeedPass with new interface: BeginFeedPass & EndFeedPass
* Refactor Pull/Push Sparse Function in box_wrapper.
* Use CUDA Kernel to copy keys and copy feasign between tensor and boxps struct.
* Cache copied keys in pull sparse in order to reuse it in push period.
5 years ago
9e29d3ebed
【OpPorting Example】DEMO OF FIX COMPILE&RUNTIME LOD_EQUALITY ( #22460 )
5 years ago
2235ee1a5e
multi-loss optimization by adding a DownpourOpt worker ( #22025 )
...
* update
* update test=develop
* update compile set test=develop
* update compile set test=develop
* update test=develop
* update test=develop
* update test=develop
* update compile setting test=develop
* update compile setting test=develop
* update run demo test=develop
* update test=develop
* update test=develop
* fix test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update format test=develop
* update format test=develop
* update style test=develop
* update style test=develop
* change style test=develop
* change style test=develop
* change style test=develop
* add dataset unittest test=develop
* update test=develop
* update for record test=develop
* udpate style for record test=develop
* update for record test=develop
* update for record test=develop
* update for record test=develop
* fix format test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
* update test=develop
5 years ago
54970444ce
Improve transpose performance with tile sm copy, test=develop ( #22311 )
...
* Refine code, fix select tile error,test=develop
* Refine element type and some comments, test=develop
* Refine comments and gpu utils, test=develop
* Remove some useless condition
* Refine floor and ceil, test=develop
* refine for loop. test=develop
Signed-off-by: zhaoyuchen <zhaoyuchen01@baidu.com>
5 years ago
a90fa54092
Compile without nccl deps. [1/2] ( #22509 )
...
支持不依赖nccl进行编译。[1/2]
多卡下，如果没有打开WITH_NCCL开关编译，多卡不能通信，则只能选择一张卡使用。
Co-authored-by: 石晓伟 <39303645+Shixiaowei02@users.noreply.github.com>
5 years ago
3a59a7a11f
Make assign op support LoDTensorArray and modify while_loop API ( #22309 )
...
This PR makes assign op support LoDTensorArray and enable the loop_vars in
while_loop to support tuple or list.
5 years ago
54a325a52f
[Refine Paddle-TRT INT8]: Support PaddleSlim's Resnet50, Mobilenetv1, Yolov3 models for Inference. ( #22483 )
...
* add int8 op teller for trt.
* refine trt int8
* add int8 op teller for trt.
test=develop
5 years ago
5739eeb9fa
add cp27-cp27m-gcc82 and cp27-cp27mu-gcc82 branch to support gcc8.2 compile for paddle, test=develop ( #22504 )
5 years ago
de009152a7
Compile without nccl deps. [2/2] ( #22484 )
...
Compile without nccl deps. [1/2]
Co-authored-by: 石晓伟 <39303645+Shixiaowei02@users.noreply.github.com>
5 years ago
4b2227e958
Fix dismatch of std::max's arguments type on windows. ( #22507 )
...
test=develop
5 years ago
870f465887
fix test_fusion_seqpool_concat lod level between compile and runtime ( #22488 )
5 years ago
a61d09527b
Fix the integer overflow problem of sequence2batch ( #22479 )
...
Fix the integer overflow problem in the op of sequence2batch, change the int32_t to size_t，
In the /paddle/fluid/operators/math/sequence2batch.h#L122.
5 years ago
197913ebe1
Add weight quantization in post_training_quanzitaion ( #22445 )
...
* support weight quantization in post_training_quanzitaion, test=develop
* add test for weight quantization, test=develop
5 years ago
dcfb603897
Enable the detection of subgraph composed of grad ops ( #21223 )
...
* Add the first implememtation of fusion_group op #19621 (#3 )
* Add the dynamic load of nvrtc, and support runtime compiling of CUDA kernel using nvrtc.
test=develop
* Call CUDA driver api to launch the kernel compiled by nvrtc.
test=develop
* Disable for mac and windows.
test=develop
* Refine the codes to support manually specified num_threads and workload_per_thread.
test=develop
* Refine the CUDA kernel to support large dims.
test=develop
* Add DeviceCodePool to manage all device codes.
* Add the first implementation fusion_group op.
* Add unit-test for fusion_group op.
* Add the check of result.
* Add the check of nvrtc in unit-test.
test=develop
* Add comment to explain the inputs, outputs and features of fusion_group op.
test=develop
* Disable fusion_group op for mac and windows.
test=develop
* Make the compiling of device code return status instead of hanging up.
test=develop
* Add the check of whether there is CUDA driver library, and do not core dump when failing to call the CUDA driver API.
* Unify fusion_group_op's input and output names.
test=develop
* Add the check of CUDA driver library in unittest.
test=develop
* Enable generating code for a given subgraph. #21126 (#4 )
* Enable generating code for a given subgraph.
* Support sorting the subgraph.
* Remove the rearange of expressions because we use the sorted subgraph directly.
* Enable generating code for a subgraph which is composed of grad ops.
* Use expression information to check the accuracy in unittest.
* Separate load and store from computation expressions.
test=develop
* Improve the loading statements in generated codes.
test=develop
* Remove unused arguments from formal list.
test=develop
* Enable the detection of subgraph of grad ops.
* Generate code for detected subgraph in fusion_group_pass.
* Add an option in BuildStrategy to enable fusion_group_pass and add unittest.
test=develop
* Fix a bug when checking whether the shape of all inputs are the same.
* Add debug information.
* Remove subgraph_detector from inference/analysis to the common framework/ir directory. (#5 )
test=develop
* Call subgraph_detector in fusion_group pass.
test=develop
* Disable fusion_group when WITH_GPU is OFF.
test=develop
* Refine all PADDLE_ENFORCE message.
test=develop
* Fix the case that some inputs are not defined in grad ops, and set op_role for fused op.
test=develop
* Follow review comments.
test=develop
5 years ago
7c9ce097f1
refine reshape_op shape error message ( #22480 )
...
test=develop
5 years ago
2b1386b2b2
optimize performance of interpolate op ( #22436 )
...
* optimize interpolate op, test=develop
5 years ago
77dd0d97bb
use enum class to replace the usage of enum in some condition test=develop ( #22464 )
5 years ago
44b45b9f07
Correct the use of DeviceContext in unittest sequence_pooling_test and sequence_padding_test ( #22456 )
...
* Add log in memory::Copy for debug purpose.
* Change to use context in DeviceContextPool directly in sequence_pooling_test, instead to new one.
* Change to use context in DeviceContextPool directly in sequence_padding_test, instead to new one.
test=develop
* Change the type of second_dim from size_t to int64_t.
test=develop
5 years ago
17f2c0899f
Add dequant-scale squash ( #22409 )
...
* Add dequant scale squash
test=develop
* Correct dequant-scale squash test
test=develop
5 years ago
9c4deedbc2
update readme of imdb training demo ( #22455 )
...
* update readme
* test=develop
5 years ago
ceda0b9b1a
[Fix BUG]: Core when multi thread + clone + paddle-trt ( #22442 )
...
* add mutex for trt engine
test=develop
* add the test for copy_to_cpu
test=develop
5 years ago
7bc4b09500
add WITH_NCCL option for cmake. ( #22384 )
...
cmake选项中添加了WITH_NCCL，显示指定是否编译NCCL的部分代码，WITH_NCCL默认打开，但如果WITH_GPU为OFF，则关闭WITH_NCCL
添加了PADDLE_WITH_NCCL定义
单机单卡能够关闭NCCL编译，多卡的话需要默认打开NCCL，如果关闭NCCL，则只能使用单卡
Co-authored-by: 石晓伟 <39303645+Shixiaowei02@users.noreply.github.com>
5 years ago
943cb8c664
fix sigmoid cudnn bug ( #22439 )
...
* Sigmoid bug fix, test=develop
* fix code format
test=develop
Co-authored-by: Manjunath Bhat <manjunathbhat9920@gmail.com>
5 years ago
d51ffe860a
fix copy table bug ( #22432 )
...
* fix copy table bug of lost some feasign
* test=develop
5 years ago
822e5b36ec
Support int16 for Tensor ( #22423 )
...
* add int16 support, test=develop
* add test, test=develop
* fix typo, test=develop
* fix dtype error in slice, test=develop
5 years ago
e1b0d7cbb1
remove anakin from code, test=develop ( #22420 )
5 years ago
0404e7a985
Update the precision of pad, pad2d, pad_constant_like's unit tests from fp32 to fp64 ( #22394 )
...
* update the ut precision of pad pad2d pad_constant_like from fp32 to fp64, test=develop
5 years ago
371f377bea
add GeneralRoleMaker ( #22295 )
...
* add GeneralRoleMaker which is for general usage
* test=develop
5 years ago
269db0d1d1
[DNNL] Fix accuracy in INT8 FC ( #22404 )
...
* Enable quantize to reorder to nchw as well
* Correct FC MKL-DNN input dim requirements to accept 3D
* Improve DNNL FC format, error and 3D input handling
test=develop
* Improve error checking in FC
test=develop
* Improve PADDLE_ENFORCE messages in fc-related files
* Remove data layout attribute from obligatory pass args
test=develop
* Fix message in fc_mkldnn_pass to be logically correct
test=develop
5 years ago
fb3086fd57
[UT coverage]Remove unnecessary transpose op registration ( #22402 )
5 years ago
ade5022681
[UT Coverage]Improve sum_mkldnn_op line coverage ( #22275 )
5 years ago
3099d9d47c
Restore requantize squash ( #22399 )
5 years ago
92462e948d
improve elementwise_add_mkldnn_op test code coverage ( #22359 )
5 years ago
20f30dd604
add benchmark flag for conv_transpose ( #22389 )
5 years ago
b96c7c9a7a
polish code, test=develop ( #22380 )
...
remove unnecessary template.
5 years ago
8f36c39537
Fix GEO-SGD init & send Bug ( #22375 )
...
* test=develop, fix geo Send & Init
5 years ago
c6f888e5a5
update unittest accuracy to float64 for relu, prelu, maxout ( #22273 )
5 years ago
0d8b222b79
Optimize the depthwise op test=develop ( #22265 )
5 years ago
aaa4fe491a
use function instead of lambda, test=develop ( #22348 )
...
* use function instead of lambda, test=develop
* follow comments, test=develop
5 years ago
e7a9f6bbb7
[Bugfix] Preserve shape in inpalce operators ( #22360 )
5 years ago
2d20869c94
Fix infer_shape in compling for elementwise_op ( #22291 )
5 years ago
b7cac50b64
Implement a common python unittest to test the ir passes. ( #22209 )
...
* Implement a common python unittest to test the ir passes.
test=develop
* Save the results in np.array and support to startup on CPU.
test=develop
* Fix the unittest.
test=develop
* Add check_program to check whether the optimized program is different from the origin one.
test=develop
* Remove the inferface all_ops.
test=develop
* Add exception test in pass_test.
test=develop
5 years ago
82bc814a57
integrated HALF_ASYNC to communicator ( #21869 )
...
* add half_async in the communicator
* fix DistributedStrategy
5 years ago
1e932eccfa
remove unused code test=develop ( #22327 )
5 years ago
3e5744aa65
Remove unused inputs for some operators ( #22284 )
...
* remove unused inputs, test=develop
* remove unused inputs, test=develop
* update dtype, test=develop
* remove unused inputs, test=develop
* update op_use_default_grad_op_maker, tese=develop
* resolve conflicts, test=develop
* follow comments, test=develop
* update center_loss_grad, test=develop
5 years ago
805328e13b
fix typo in error message ( #22312 )
5 years ago
895f8da7d6
change std::cout to log(INFO), vlog ( #22316 )
5 years ago
8cb04664b9
revert paddle_fluid.map, test=develop ( #22236 )
5 years ago
35efbe6d95
Speeding up dygraph DataLoader with multiprocessing ( #21762 )
...
* add multiprocess for dygraph data loader, test=develop
* polish code & add safe gurad, test=develop
* refactor dygraph dataloader & add signal handler, test=develop
* fix member initializer compile error on ci, test=develop
* fix member initializer compile error one more, test=develop
* remove useless config, test=develop
* skip windows incompatible problem, test=develop
* add unittest for coverage, test=coverage
* add more exception unittest case, test=develop
* deal with signal handler coverage, test=develop
* polish code & add signal handler tests, test=develop
* deal with coverage ci problem, test=develop
* split data loader test & coverage ci fix, test=develop
* remove test_imperative_data_loader_with_exception, test=develop
* remove singal process except test case, test=develop
* add exception tests again & remove sample list test, test=develop
* split normal and exception unittests to diff class, test=develop
* polish doc for use_multiprocess effect in static mode, test=develop
5 years ago
9435533adf
remove op_use_default_grad_op_maker.spec, test=develop, test=document_fix ( #22300 )
5 years ago
7b76a76495
fix the conda build confilict test=develop ( #22279 )
5 years ago
5e601a92ad
polish grad op check ( #22290 )
...
* polish grad op check, test=develop, test=document_fix
* keep op_use_default_grad_maker.spec to avoid conflict, test=develop, test=document_fix
5 years ago
faba4b116a
Remove disable flag in test_fsp_op.py ( #22171 )
...
* fix fsp_op, test=develop
* fix fsp grad op maker, test=develop
* update op_use_default_grad_op_maker.spec, test=develop
5 years ago
e40cfb1010
fix the bug of assert_is_op_output. test=develop ( #22262 )
5 years ago
d3a6647372
improve placement pass tests code coverage ( #22197 )
5 years ago
07afc29e90
Make api.cc malloc consistent with paddle_api.h for PaddleBuf ( #22255 )
5 years ago
4f1da4adcb
remove the useless third_party library from C++ inference library ( #22021 )
...
* remove the useless third_party library from C++ inference library
* revert removing the install directory
5 years ago
549e6de7ac
faster build by reduce by-product, reduce linking library and fix compile warning of std=c++11 ( #22164 )
5 years ago
songyouwei
Zhaolong Xing
Jacek Czaja
Yi Liu
wangchaochaohu
Leo Chen
GaoWei8
石晓伟
Zeng Jinle
zhaoyuchen2018
Wojciech Uss
Wilber
Tao Luo
xujiaqi01
danleifeng
liym27
Liufang Sang
Pei Yang
Adam
Yiqun Liu
hutuxian
Sylwester Fraczek
Zhang Ting
Feiyu Chan
yaoxuefeng
chengjuntao
wawltor
guofei
liu zhengxi
tangwei12
mapingshuo
Michał Gallus
Chen Weihang
hong
Zhen Wang
zhongpu
tianshuo78520a
Kaipeng Deng
wangguanzhong
FlyingQianMM
zhouwei25
zhangchunle
dyning
ShenLiang
Yibing Liu
flame
Huihuang Zheng
lidanqing
123malin
Jiawei Wang
Steffy-zxf
Guo Sheng
Double_V
huzhiqiang
Zhong Hui
cc
LielinJiang
joanna.wozna.intel
ceci3
Chengmo
zhupengyang
qingqing01
Bai Yifan
silingtong123