d0f0a2520c
test Optimizer in dygraph ( #21949 )
...
* test Optimizer in dygraph, test=develop
* add optest for Optimizer in dygraph, test=develop
* fix adagrad optimizer, test=develop
* fix dpsgd optimizer, test=develop
* fix test_optimizer.py, test=develop
* fix dpsgd optimizer, this op only support cpu, test=develop
* add optest for optimizer, test=develop
* add description for dpsgd, test=develop
* add rmsprop to white_list in unused_var_check.cc, test=develop
* polish code style, test=develop
* polish code style, test=develop
* delete seed attribute for DpsgdOptimizer, test=develop
* change testing to debugging, test=develop
6 years ago
83037e55ca
named_sublayers and named_parameters ( #21868 )
...
* use snake_cased Layer name
* add named_parameters and named_sublayers api
* add include_sublayers param, add unittest
test=develop
* fix named unittests
test=develop
* fix unittest
test=develop
* add api docs
test=develop
* arg fix
test=develop
* reserve rnn_impl name_scope for static graph
test=develop
* fix load static param
test=develop
* fix load static param
test=develop
6 years ago
7d10edc5ee
add clear_gradients for Optimizer and add clear_gradients api description ( #21948 )
...
* add clear_gradients for Optimizer, add api description, test=develop
* fix optest for optimizer's clear_gradient interface, test=develop
* add sample code, test=develop
* polish sample code, test=develop
6 years ago
51cb918a05
update layers used in transformer dygraph model, test=develop ( #22051 )
...
* update layers, test=develop
* update layers for resnet, test=develop
* fix is_test attr, test=develop
* update cycle_gan, test=develop
* update reinforcement_learning, test=develop
* update ocr, test=develop
* fix bug, test=develop
6 years ago
1ae3f47f59
update PRelu arg ( #21946 )
...
* update PRelu arg
test=develop
* fix unittests
test=develop
* fix element mode alpha shape
test=develop
* split channel_or_input_shape arg
test=develop
* fix unittest
test=develop
6 years ago
8b29fec7ad
open test dygraph mnist fp16, test=develop ( #22163 )
6 years ago
e587bc62f9
remove useless code, test=develop ( #22111 )
6 years ago
c3bcd3c1e2
fix dygraph non zero gpu bug, test=develop ( #22165 )
6 years ago
cfb0c12ef9
fix grad diff over limit when the data type of input is double,test=develop ( #22161 )
6 years ago
ccac636d5e
add example for reshape op ( #22129 )
...
* add example for reshape op
6 years ago
5e07db15e6
fix init scaling value test=develop ( #22145 )
6 years ago
cf475f95df
Remove FC in dygraph, modify FC to Linear in sample code ( #22082 )
...
* modify fc to linear in sample code, test=develop
* remove FC, test=develop
* remove warnings, test=develop
* drop fluid/imperative/README.md , test=develop
* change fc to linear, test=develop
* polish code style, test=develop
6 years ago
6ea3809143
Support prroi_pool_op with Tensor and LoDTensor rois ( #20649 )
...
1. Add a new input named batch_roi_nums for prroi_pool_op. batch_roi_nums includes the number of roi for each image in batch when rois is Tensor. This information is saved in rois's lod when rois is LoDTensor.
2. add grad check to prroi_pool_op and solve unnormal X grad diff in CPU.
6 years ago
bde7ee97f7
Use unified threshold in all op test ( #21814 )
...
* modify test framework, test=develop
* fix test_fsp_op, test=develop
* up, test=develop
* up, test=develop
* up, test=develop
* add check for threshold, test=develop
* up, test=develop
* up, test=develop
* up, test=develop
* up, test=develop
* up, test=develop
* up, test=develop
* up, test=develop
* update, test=develop
* update, test=develop
* update, test=develop
6 years ago
d0406193e3
LayerList add sublayer none check ( #22135 )
...
test=develop
6 years ago
f2f913073f
all cases use large shape ( #22102 )
...
enhanced ops: acos, asin, brelu, ceil, cos, elu, floor, hard_shrink,
hard_sigmoid, hard_swish, relu6, round, rsqrt, sin, softshrink,
cos_sim, cross_entropy, im2squence, kldiv_loss
6 years ago
9f7d90d203
Load inference enhance ( #21919 )
...
* enhance load interface; test=develop
* add uni test and add comment; test=develop
* fix converage; test=develop
* use path.joint replace "/"; test=develop
* windows debug; test=develop
* fix window unitest error; test=develop
* fix commet error; test=develop
* add model shuffix check; test=develop
* fix example error; test=develop
6 years ago
2f49cf70f7
Support dygraph structuerd name ( #21930 )
...
* support dygraph structured name; test=develop
* add load static param unitest and fix save load; test=develop
* fix varBase import error; test=develop
* fix unitest error; test=develop
* add comment for parameter; test=develop
* fix uni test error; test=develop
* change parallel se-resnet; test=develop
* fix dygraph se resnext parallel test; test=develop
* remove useless code; test=develop
* remove useless code; test=develop
6 years ago
b9a6354710
Solve elu unitest fail ( #22123 )
...
* set esp as 1e-6 to solve elu unitest fail,test=develop
6 years ago
418abc92f4
Update pyramid related OP ( #21372 )
...
* add special way to add distribute vars， Update Pyramid hash op
6 years ago
3b84584eb4
fix Variable's gradient api in framework.py, test=develop ( #21577 )
...
* fix Variable's gradient api in framework.py, test=develop
* remove namescope, test=develop
6 years ago
434f7b9c0f
Fix the global_step & continuous applying error in EMA ( #22090 )
...
* Fix the global_step & continuous applying error in EMA
test=develop
* Fix for step 0 & add unit test, test=develop
6 years ago
5de6a1913b
fix test_bilinear_tensor_product_op timeout ( #22120 )
6 years ago
de56887bef
add Note in the doc of old control flow ops. test=develop,test=document_fix ( #22108 )
6 years ago
5c33919328
Fix grad clip ( #21784 )
...
* fix grad clip， clip op belongs to Backward op when running in Parameter Server mode.
6 years ago
14aebc7a95
add erf op ( #21785 )
...
* add erf op and python interface.
* add fp16 support for erf op.
* add unitests for erf op and its python interface.
6 years ago
f385c34107
fix fleet collective api run on cpu, test=develop ( #22064 )
6 years ago
cce9af0e2b
all cases use large shape ( #22106 )
...
enhanced ops: conv2d, conv3d
elementwise_pow: change to a reasonable shape
6 years ago
fab4b0765a
support elu_op double grad ( #21822 )
...
* support elu activation double grad,test=develop
* delete the code commit in .cc,test=develop
* fix relu test unpass, test=develop
* add elu double grad kernel and unit test
* add caculate dX in elu double grad functor, test=develop
* update the commit code,test=develop
6 years ago
b0b27ff699
[MKL-DNN] Conv grad and Batch Norm grad NHWC support ( #22088 )
6 years ago
1ce6ab9cc6
Refine doc of data based on requirement. ( #22098 )
6 years ago
dd4361568e
Add ParallelExecutor Test for Cond API and Fix PE Checks Shape Bug ( #22029 )
6 years ago
2a47cc5fc4
dygraph LayerList container ( #21734 )
...
* add dygraph LayerList
* add unittest
test=develop
* add newline
test=develop
* typo fix
test=develop
6 years ago
288b1e75de
Fix segmentation fault bug in reduce ops ( #22079 )
...
* fix dim=[] bug in reduce ops
6 years ago
385ed8a5e1
fix warning in test_install_check.py, test=develop ( #22087 )
6 years ago
370a8e9126
Remove unit tests that are no longer maintained ( #22081 )
6 years ago
e519a1fc82
add bigger shape test example for elementwise_pow OP ( #22063 )
6 years ago
8b3ef2a587
all cases use large shape ( #22091 )
...
enhanced ops: unsqueeze, squeeze2, strided_slice, unsqueeze,
unsqueeze2, var_conv_2d, spectral_norm, slice, match_matrix_tensor,
nce, pad, pad_constant_like, filter_by_instag
6 years ago
85ba527522
all cases use large shape ( #22065 )
...
affine_channel, affine_grid, bilinear_interp, bilinear_tensor_product,
clip, crop, crop_tensor, cumsum, data_norm, expand, expand_as
6 years ago
7fb817d447
add distributed_strategy ( #21710 )
...
* add distributed_strategy
6 years ago
ad8a9cb82c
[MKL-DNN] Pool & LRN Grad Ops NHWC support ( #21747 )
6 years ago
e1d666fb8b
Hotfix to MKL-DNN pool2d tests ( #22086 )
6 years ago
bf978fffe5
all cases use large shape ( #22084 )
...
flatten, flatten2, gather, gather_nd, group_norm, gru,
gru_unit, reduce_max, reduce_min, reduce_mean, reduce_sum,
reshape, scatter_nd_add, sequence_expand, sequence_expand_as,
sequence_pad
6 years ago
7d8d45991a
control flow: support optimizer called ( #21851 )
...
* append optimize op in the grad block of current block if current block is in control flow. test=develop
* add conditional grad op when optimizer used in control flow. test=develop
* add comment and modify typo. test=develop
* fix append_backward to support control flow. test=develop
* add test. test=develop
* fix copy_var_to_parent_block and conditional_block_grad. test=develop
* fix bug: revert to append conditional_block_grad vars to sub grad block. test=develop
* fix bug: revert to assign var to parent block even if var already is in parent block
* fix bug: consider outputs is empty. test=develop
* move _rename_grad_ out. test=develop
* modify code according to reviews from Huihuang. test=develop
* modify code according to reviews from Jinle. test=develop
6 years ago
12b2b4b1a1
Add gradient check to reduce_max and reduce_min ops ( #21985 )
6 years ago
3fab86ffa6
add package for centerloss ( #21806 )
6 years ago
1c39efb783
Enable test conv2d ngraph ( #22074 )
6 years ago
c112b645c4
Update MKL-DNN to 1.1 ( #21754 )
6 years ago
3f653c8323
register NoNeedBufferVarsInference for max_pool_grad_op, test=develop ( #22055 )
...
* fix test_conv2d_ngraph for grad diff, test=develop
* register NoNeedBufferVarsInference for max_pool_grad_op, test=develop
* refine error message, test=develop
* fix numpy, test=develop
* disable test conv2d_ngraph_op, test=develop
Co-authored-by: Zhang Ting <709968123@qq.com>
6 years ago
Liufang Sang
5b8837897d
add shape description in accuracy api doc ( #22011 )
...
* add shape description in accuracy api doc test=develop
* fix details test=develop
* fix details test=develop
6 years ago
6192108408
[DNNL] 3D Fully-Connected ( #21746 )
6 years ago
c1fea3e3e2
Fix example in log_loss doc ( #22052 )
...
* Fix example in log_loss doc, test=develop
* Use none for unknown dim, test=develop, test=document_fix
6 years ago
f5f65a7e8a
OP unit test specification: compiletime-runtime consistency check ( #22006 )
6 years ago
aa2ed0dcc6
fix generate_proposal_labesl op ( #21793 )
...
* test=develop fix generate_proposal_labesl op
6 years ago
81030125a3
convert input vars' dtype for range op ( #22028 )
...
* convert dtype of vars for range op, test=develop
6 years ago
95d79b6d00
update error log for batch_norm_grad ( #22017 )
...
* update error information about batch_norm_grad
* update bn,test=develop
6 years ago
985e4bae5e
fix unittest for spp op, test=develop ( #22030 )
...
fix unittest for spp op
6 years ago
56414c7daf
move private weight fields to public ones ( #21982 )
...
* move private weight properties to public ones
test=develop
* revert changes to FC
test=develop
* fix unittest
test=develop
* fix unittest
test=develop
* fix coverage
test=develop
* fix merged dev
test=develop
* bug fix
test=develop
6 years ago
b852ef7375
add no_check_set check for op unittests ( #21611 )
6 years ago
c7b03d308c
open test_parallel_executor_fetch_feed for PR_Windows_CI. test=develop ( #22020 )
6 years ago
0aab2578f2
fix scipy bug ( #22024 )
6 years ago
9a2204ee1f
Uniform append_backward & gradients parameter_list type to Variable ( #21938 )
...
* update doc, test=develop
* fix related unittests, test=develop
* fix str incompatible error, test=develop
6 years ago
6b4c33ee30
fix decay param in DecayAdagrad test=develop ( #22026 )
6 years ago
835ba4ec81
add is_empty_grad_op for OpTest ( #22022 )
...
* add is_empty_grad_op for OpTest, test=develop
6 years ago
34e26c9a7d
remove @unittest.skip in test_parallel_executor_mnist.py ( #22019 )
6 years ago
b68f4e006d
Fix the doc of scatter_nd ( #22015 )
...
fix the doc of scatter
6 years ago
3ec289a6a3
fix sync_batch_norm hang in fleet ( #21838 )
6 years ago
a0b533765e
increase data shape in topk_avg_pooling unittest ( #22010 )
6 years ago
86c40e207d
Expose fluid.dygraph.TracedLayer API ( #21518 )
...
* expost fluid.dygraph.TracedLayer apis, test=develop
* polish doc, test=develop
* follow comments, test=develop, test=document_fix
* follow comments, test=develop
* remove save_inference_model return value, test=develop
6 years ago
911eef4319
enhance checking validity of apply_gradient ( #21870 )
...
* enhance checking validity of apply_gradient test=develop
* refactor code test=develop
* stash
* add adamOptimizer test=develop
* add adamOptimizer test=develop
6 years ago
33f1306753
update layers used in mnist dygraph model, test=develop ( #21947 )
...
* update layers used in mnist dygraph model, test=develop
* fix import issue, test=develop
* add dygraph utils, test=develop
* add unittest, test=develop
6 years ago
c3527f5526
Annotate NormOP test to skip grad check ( #21894 )
6 years ago
fa8bab9cbc
add fp64 check_grad for conv2d, conv3d ( #21931 )
...
* test=develop add fp64 check_grad for conv2d, conv3d
6 years ago
cf7b325389
fix conv2d_transpose op unittest ( #21927 )
6 years ago
f4013ecb3b
enable kldiv_loss_op unittests ( #21990 )
6 years ago
c15d4208b9
fix no grad check for pool_max & spectral_norm ( #21971 )
...
* fix no grad check for pool_max & spectral_norm. test=develop
6 years ago
e0d8b8f5c0
use large input size in test_add_position_encoding_op.py ( #21962 )
...
* use large input size in test_add_position_encoding_op.py, test=develop
* revert shape_white_list, test=develop
6 years ago
362bfec52e
add xly py35 dockerfile ( #21861 )
...
* add xly py35 dockerfile
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test dockerfile
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35
* test=py35,test=develop,test=inference,test=document_fix
6 years ago
ba51043f7e
remove parameter block in to_variable, test=develop ( #21933 )
6 years ago
cee2ccb078
add shuffle batch op ( #21674 )
...
* add shuffle batch op, test=develop, test=document_preview
* fix size_t conflict and check_output test=develop, test=document_preview
* fix bug test=develop, test=document_preview
* add unittest of shuffle_batch layer test=develop, test=document_preview
* fix py coverage and op input type, test=develop, test=document_preview
* fix py coverage, test=develop
* fix en doc, test=develop
* move to contrib test=develop
* add unique_name test=develop
* invoke shuffle_batch in contrib.layers test=develop
6 years ago
dca075839b
remove params in Tracer object (in dygraph) ( #20815 )
...
* remove params in Tracer object, test=develop
* Repair failed optest, test=develop
* remove build_once & name_scope (Conv2D)
test=develop
* fix unittest
test=develop
* Conv2DTranspose
* Conv3D & Conv3DTranspose
test=develop
* Pool2D & BatchNorm
* Embedding
* LayerNorm
* GRUUnit & NCE
* PRelu
* BilinearTensorProduct
* GroupNorm & SpectralNorm
* TreeConv
test=develop
* fix LayerNorm in transformer unnittest
test=develop
* disable LayerNorm or BatchNorm in multicard
test=develop
* refine Layer.create_parameter api
test=develop
* refine LayerNorm, remove begin_norm_axis param, add normed shape check
test=develop
* LayerNorm bug fix
test=develop
* fix optest,test=develop
* fix optest, test=develop
* fix optest for pass parameter_list when constructing an Optimizer class instance, test=develop
* polish code for better code style, test=develop
* fix se_resnext optest, test=develop
* polish code for better code style, test=develop
Co-authored-by: songyouwei <youwei0314@gmail.com>
6 years ago
c3e1954918
make reverse op support negative axis ( #21925 )
...
* make reverse op support negative axis
6 years ago
911e4c7db6
Fix the bug in sequence_scatter unitest ( #21960 )
...
* test=develop
* test=develop
* test=develop
* change the check_shape_white_list in web
* test=develop
* test=develop
6 years ago
2687f2afec
fix input shape of test_log_loss_op ( #21953 )
6 years ago
b1ec1d54cf
Delete check grad for int, collect op for now ( #21922 )
6 years ago
2df4be5d35
Fix openblas bug to support compile on windows when WITH_MKL=OFF ( #21902 )
...
* Fix openblas to support compile on Windows when WITH_MKL=OFF
6 years ago
27decacb8a
fix aucop stat shape ( #21846 )
...
* fix stat shape back in global auc scenario
* add UT to cover global auc
6 years ago
0d61653c8e
Revert to run distributed unittest serially, test=develop ( #21924 )
6 years ago
cff7a49856
Add ParameterList ( #21056 )
...
* add ParameterList container
* add unittest
test=develop
* fix container, rebase remove build_once
test=develop
* add parameters None check
test=develop
* update Sequential unittest
test=develop
* use Linear in Sequential sample code
test=develop
* fix Sequential arg type doc
test=develop
6 years ago
8439384e25
Move sequence op unittest files to a separate folder sequence ( #21842 )
...
* move sequence op unittests to a separate folder
test=develop
* add missing CMakeList file
test=develop
* fix relative path import
test=develop
* fix relative import
test=develop
* use sys.path.append
test=develop
6 years ago
2690ddb558
Use skip_check_grad_ci to skip grad check ( #21907 )
6 years ago
1f3029cb54
ci check for check_shape_white_list.py ( #21859 )
6 years ago
51a86d2b6b
Optimize adam speed ( #21777 )
...
* optimize adam speed by removing _finish_update test=develop
* fix SparseAdamFunctor param list test=develop
* Remove scale_op in expect_list of adam_op test=develop
* fix test optimizer loss assert error test=develop
* fix test optimizer loss assert error test=develop
* modify PADDLE_ENFORCE usage test=develop
* fix op_type in lamb_op.cc test=develop
* fix errors ostream format bug test=develop
* add betaPowOut in ngraph op test=develop
* fix ngraph::op api for gcc8 test=develop
* clean code test=develop
* modify struct into class test=develop
* remove code of beta1Tensor in lamb_op test=develop
6 years ago
310edc0d0c
Update layers used in ptb model to use auto-generated op functions in dygraph mode ( #21724 )
...
* update layers, test=develop
* fix input numpy, test=develop
* fix bugs, test=develop
* follow commments, test=develop
* update getitem, test=develop
6 years ago
f17bd1786a
add decorator skip_check_grad_ci for test_group_norm ( #21910 )
...
* add decorator skip_check_grad_ci for test_group_norm test=develop
6 years ago
feabed1309
Modify op test ( #21835 )
...
* remove return in op_test teardown, test=develop
* add check_grad for fp64 and other type
* update white list
6 years ago
53c16eabe4
test the compatibility between execution time and compiling time ( #21511 )
...
* test the equality between running-time and compiling time
* add white_list for compile_vs_runtime check test=develop
* add check of new-registered op on compile&runtime-lodEquality test=develop
* test=develop
* test=develop
* update white-list test=develop
* merge with new code test=develop
* fix code style test=develop
* test=devleop
* test=develop
* test=develop
* test=develop
* add more ops to white_list test=develop
* test=develop
* test=develop
* test=develop
* test=develop
* test=develop
* fix code_style: add comments test=develop
* test=develop
6 years ago
9dff56e8e2
change qat_performance with mobilenet, change batch_size of qat2_resnet50 ( #21895 )
...
test=develop
6 years ago
5b282a8157
skip check_grad of TestHSigmoidOpSparse ( #21888 )
6 years ago
6b9fbcf3ad
Update iou_similarity op to support non-normalized bbox ( #21671 )
...
Update iou_similarity op to support non-normalized bbox
6 years ago
d053561d84
use large input size in tree_conv and teacher_student_sigmoid_loss ( #21896 )
...
test=develop
6 years ago
46f9184aff
Modify the while_loop API ( #21844 )
6 years ago
7689b6aaa4
Fix default label dim of label_smooth_op. test=develop ( #21862 )
6 years ago
c63ce41102
resolve #21799 ( #21872 )
...
fix flip comments, test=develop
closes:#21799
6 years ago
52a61edaad
rename test_dist_mnist & run dist parallel, test=develop ( #21874 )
6 years ago
879e3074ea
fix the unittest of test_check_import_scipy on V100 ( #21853 )
6 years ago
a25418b32c
Set dtype of compare op's output to bool ( #21864 )
...
* add unittests, test=develop
* set dtype of compare op to bool, test=develop
6 years ago
934d99867a
add selected rows supported in framework ( #21808 )
...
* add selected rows supported in framework
6 years ago
855ed5fb3d
revert parallel test=develop ( #21856 )
6 years ago
25e765a4fe
Copy stop_gradient property of variable in clone ( #21825 )
...
* copy stop_gradient property in clone, test=develop
* fix pruning issue, test=develop
* fix kLookupTablePath issue, test=develop
* sync program, test=develop
* copy subblocks, test=develop
* add unittests, test=develop
6 years ago
957e182a90
revert a+b axis change; test=develop ( #21821 )
6 years ago
6624ea7673
skip test_fetch_handler, will investigate it later, test=develop ( #21845 )
6 years ago
2b941736f3
fix softmax_with_cross_entropy_fix bug, test=develop ( #21810 )
6 years ago
f4379a9149
Update the precision of some op tests from fp32 to fp64 ( #21847 )
6 years ago
538c848958
add decorator skip_check_grad_ci ( #21836 )
6 years ago
bf9c5de752
Update test precision from fp32 to fp64 ( #21848 )
6 years ago
2b1f08c76c
Remove self-set accuracy parameters of op tests: max_relative_error ( #21815 )
6 years ago
9bb1e3ea16
add shape_check for op tests ( #21391 )
6 years ago
63a5fb4cac
tmp fix depthwise_conv2d in dygraph ( #21840 )
...
test=develop
6 years ago
5bb9deb335
Update the precision of some op tests from fp32 to fp64 ( #21774 )
6 years ago
a86f11b5f5
Speed GEO dense calc & communication ( #21579 )
...
* test=develop, speed dense calc & communication
6 years ago
666c3bb9b0
handle multi-inputs with empty inputs for mkldnn_concat_op ( #21827 )
...
test=develop
6 years ago
a9af87edbc
Remove self-set accuracy parameters of op tests: max_relative_error ( #21823 )
...
* Remove self-set accuracy parameters of op tests: max_relative_error
test=develop
* fix error
test=develop
6 years ago
e53d5967c9
Remove self-set accuracy parameters of op tests: max_relative_error ( #21816 )
...
* Remove self-set accuracy parameters of op tests: max_relative_erro
test=develop
* fix error
test=develop
6 years ago
1395828408
Add dygraph Linear layer ( #21265 )
...
* add Linear layer
test=develop
* update unittest for coverage
test=develop
6 years ago
d683b65b1a
Remove self-set accuracy parameters of op tests: atol ( #21711 )
...
* Remove self-set accuracy parameters of op tests:atol
test=develop
* keep smaller parameters
test=develop
* fix error
test=develop
6 years ago
8b74fc4fa7
Fix post training quantization ( #21745 )
...
* fix post training quantization bug of memory constrained, support the input be different, test=develop
6 years ago
aa4d6a5d6c
Add some debug flags to auto growth allocator ( #21766 )
...
* add some debug flags to auto growth allocator, test=develop
* add comments about auto growth, test=develop
6 years ago
c50ebeac12
add comments of inplace_atol ( #21819 )
6 years ago
8b7c50f49a
Make While Op could run on GPU place and add while_loop unittest ( #21672 )
...
1. Make while_op accept GPU conditional data
2. Add more complex test cases for while_loop API
6 years ago
a3a3558dd0
Remove self-set accuracy parameters of op tests: max_relative_error ( #21817 )
...
test=develop
6 years ago
187d1c38ef
Remove self-set accuracy parameters of op tests: max_relative_error ( #21744 )
...
* Remove self-set accuracy parameters of op tests: max_relative_error
test=develop
* fix errors
test=develop
6 years ago
3c33417905
modify the method of skipping CI in distributed unittests ( #21764 )
6 years ago
fb067fa4d2
Modify test framework, test=develop ( #21789 )
...
*use dtype to determine whether check_grade is needed, and delete useless class
6 years ago
557bce77da
Fix Backward Bugs in Conditional Block ( #21809 )
...
The fixed bugs:
1. The condition sub-graph is not pruned
2. When backward graph is extremely simple, the whole backward ops are pruned.
6 years ago
eab124ba98
fix accuracy parameters of op tests ( #21813 )
...
test=develop
6 years ago
642b33564e
Update test precision from fp32 to fp64 ( #21805 )
6 years ago
04909137f5
fix lr assert, test=develop ( #21780 )
6 years ago
9894a4fb35
update test precision from fp32 to fp64, test=develop ( #21783 )
6 years ago
c96f06f2f6
add unary operator __neg__, test=develop ( #21787 )
...
adds unary operator __neg__ for VarBase in dygraph mode, and for Variable in static graph mode.
6 years ago
65c8eac9fe
Remove self-set accuracy parameters of op tests: max_relative_error ( #21741 )
...
* Remove self-set accuracy parameters of op tests: max_relative_error
test=develop
* Remove self-set accuracy parameters of op tests: max_relative_error
test=develop
* Remove self-set parameters of op tests: max_relative_error
test=develop
6 years ago
579a159a7b
Remove self-set accuracy parameters of op tests: max_relative_error ( #21739 )
...
* Remove self-set accuracy parameters of op tests: max_relative_error
test=develop
* Remove self-set accuracy parameters of op tests
test=develop
* keep smaller parameters
test=develop
6 years ago
ecb2419e5c
increase the explanation doc of py_func ( #21631 )
...
* increase example code of py_func, fix some wrong description of English API doc
6 years ago
b976ba3e83
fix unittests ( #21786 )
...
test=develop
6 years ago
73e97d39b4
add check for check_grad in Op unittests ( #21383 )
6 years ago
0677a1c1c1
Fix That conditional_block_op Doesn't Have InferShape ( #21733 )
6 years ago
ac666b8ac3
Remove self-set accuracy parameters of op tests: atol ( #21731 )
...
* Remove self-set accuracy parameters of op tests
test=develop
* add cast
test=develop
* Remove self-set accuracy parameters of op tests:atol
test=develop
* lrn_ngraph_op
test=develop
* Keep smaller parameters
test=develop
6 years ago
c924353d4f
Remove self-set accuracy parameters of op tests: max_relative_error ( #21740 )
6 years ago
f6144d8463
remove build_once & name_scope ( #21131 )
...
* remove build_once & name_scope (Conv2D)
test=develop
* fix unittest
test=develop
* Conv2DTranspose
* Conv3D & Conv3DTranspose
test=develop
* Pool2D & BatchNorm
* Embedding
* LayerNorm
* GRUUnit & NCE
* PRelu
* BilinearTensorProduct
* GroupNorm & SpectralNorm
* TreeConv
test=develop
* fix LayerNorm in transformer unnittest
test=develop
* disable LayerNorm or BatchNorm in multicard
test=develop
* refine Layer.create_parameter api
test=develop
* refine LayerNorm, remove begin_norm_axis param, add normed shape check
test=develop
* LayerNorm bug fix
test=develop
6 years ago
0fe16539ef
Fix dgc & launch tests in cpu ci ( #21759 )
6 years ago
a5a8d14414
Fix softmax cuda bug ( #21720 )
...
* Fix softmax cuda bug
* Refine multihead log and softmax logic
6 years ago
943a44492b
yolo_box OP add Attr(clip_bbox). ( #21620 )
...
* yolo_box OP add Attr(clip_bbox). test=develop
6 years ago
c4f8f3bddc
use large input shape for accuracy test ( #21758 )
...
- ngarph: elementwise_sub, elementwise_mul
- mkldnn: transpose, sum
- others: scatter_nd
test=develop
6 years ago
92f83e46aa
fix accuracy parameters of op tests ( #21729 )
...
test=develop
6 years ago
e92d113590
fix bug that tuple(Variable) is converted to list(Variable) uncorrectly ( #21687 )
6 years ago
7181afd75c
Fix elementwise_pow bug on CUDA place with integer ( #21675 )
...
* fix elementwise_pow bug on integer, test=develop
* use llrint to support elementwise_pow_grad, test=develop
* add some tests, test=develop
* revert grad functor, test=develop
6 years ago
68999b6c7a
simplify dygraph data loader code, test=develop ( #21722 )
6 years ago
9c481e12ba
Patch math method for VarBase using auto-generated op functions ( #21656 )
...
* patch math method for varbase using auto-generated op functions, test=develop
* clean code that handles batch_size, test=develop
* follow comments, test=develop
* follow comments, test=develop
* code clean, test=develop
6 years ago
aa287e19fe
remove the dependency on ssl ( #21712 )
6 years ago
1ee5ba1c58
Enhance checking on sub branch of backward ( #21582 )
...
* add backward unittest test=develop
* add sub-branch in test_backward.py test=develop
* refine code, add comment test=develop
* reconstruct TestBackward Class test=develop
* fix typo of comment test=develop
6 years ago
4570295122
fix assert in nan inf tests, test=develop ( #21742 )
6 years ago
549f24b5f1
run dist tests parallel( #21751 )
6 years ago
a6e935f424
Update op test framework ( #21599 )
...
* update op test framework
6 years ago
7c38612347
disable op test of kldiv_loss ( #21749 )
6 years ago
d528ffaa04
use large input shape for accuracy test ( #21716 )
...
affine_grid, label_smooth, spectral_norm, warpctc,
nearest_interp, data_norm, match_matrix_tensor,
var_conv_2d, fused_embedding_seq_pool
test=develop
6 years ago
a2175cfc96
Tmp fix fleet bug in py35 gcc8 CI, test=develop ( #21703 )
6 years ago
227edfe9da
use large input shape for op tests ( #21705 )
...
sequence_expand_as, squared_l2_distance, gather_nd,
center_loss, rank_loss, conv_shift, spp,
modified_huber_loss, smooth_l1_loss, multiplex,
sequence_softmax, nce, huber_loss, group_norm,
kldiv_loss, hinge_loss, expand_as
test=develop
6 years ago
d419b859c0
Add reshape int8 mkldnn op ( #21428 )
...
* Add reshape int8 op
test=develop
* Change test to CPUPlace
test=develop
* Correct tests
test=develop
6 years ago
c047e713b0
make nGraph ut py3 compatible ( #21679 )
6 years ago
3b915741b2
Fix nGraph UT for PY3 - Part I ( #21678 )
...
* fix test for PY3 test=develop
* reduce file changes test=develop
6 years ago
8a0f611b64
Rewrite check nan inf tools ( #21076 )
6 years ago
019147eb8b
use large input shape for accuracy test, ( #21693 )
...
sequence_unpad, expand, pad, pad_constant_like,
norm, bilinear_tensor_product, flatten2, im2sequence,
unpool, cos_sim, strided_slice, flatten, elementwise_min,
abs, acos
test=develop
6 years ago
5c4106023c
disable qat int8 test for now, test=develop ( #21696 )
6 years ago
9ad940fdfe
memory leak for cpu ( #21174 )
...
* add fake init for the trainer, fix large memory hold in the trainer
* do not merge recv vars from a remote endpoint, test=develop
* add recv and save op, merge slice var in one op, save memory
* remove hsigmoid with pull sparse, test=develop
6 years ago
4c987a6003
fix input shape of op tests ( #21682 )
...
* fix input shape of op tests for elementwise_sub,
gather, pad2d, transpose, softmax, scale,
elementwise_max, hierarchical_sigmoid, reshape2,
sign, squeeze, reduce_sum, sum, squeeze2,
unsqueeze, unsqueeze2, cast, reverse
test=develop
* fix cast, elementwise_mul, gather, scale, sign,
softmax, transpose
test=develop
6 years ago
f64d006622
Change several tests to inherit the right parent class, test=develop ( #21652 )
...
* change several tests to use the right parent class, test=develop
* add dtype for TestLoDTensorAndSelectedRowsOp, test=develop
6 years ago
686f0ecb6a
add `no_need_buffer_slots` interface to pybind ( #21575 )
...
* add no_need_buffer_slots interface to pybind
6 years ago
52f38877e4
Fix ci bug for deleting data files when other test is running ( #21661 )
...
* fix ci bug for deleting data files, test=develop
* update, test=develop
6 years ago
99331fa113
Fix current block in math_patch_op ( #21189 )
6 years ago
d96acc3363
Refine dygraph DataLoader implementation ( #21634 )
...
* refine dygraph dataloader & polish related code, test=develop
* refine code based review comment, test=develop
6 years ago
e2d849b989
Dropout with seed ( #21590 )
...
* add seed op
6 years ago
e81f0228df
MKL-DNN 1.0 Update ( #20162 )
...
* MKLDNN v1.0 rebase to Paddle 1.6
test=develop
* Add hacky paddle::string::to_string() implementation
* vectorize<int64-t>() -> vectorize() cleanup
test=develop
* PADDLE_ENFORCE and void_cast fixes
test=develop
* Rebase changes
test=develop
* Cosmetics
test=develop
* Delete MKL from mkldnn.cmake
test=develop
* CMake debug commands
test=develop
* Delete MKLDNN_VERBOSE and rebase fixes
test=develop
* Rebase fixes
test=develop
* Temporarily disable int8 resnet101 vgg16 and vgg19 tests
test=develop
* Add libmkldnn.so.1 to python setup
test=develop
* Add libmkldnn.so.1 to inference_lib cmake after rebase
test=develop
* Post rebase fixes + FC int8 changes
test=develop
* Fix LRN NHWC
test=develop
* Fix NHWC conv3d
test=develop
* Windows build fix + next conv3d fix
test=develop
* Fix conv2d on AVX2 machines
test=develop
6 years ago
f404157205
fix master patch when slot is dense ( #21580 )
...
* fix master patch when slot is dense
* test=develop
6 years ago
48600d7f17
Add op function generator for dygraph ( #21569 )
...
* add op function generator, test=develop
* add unittest, test=develop
* follow comments, test=develop
* fix windows compilation problem, test=develop
6 years ago
9a4dd1bc25
support float64 for GradClipByGlobalNorm in dygraph, test=develop ( #21401 )
...
* support float64 for GradClipByGlobalNorm in dygraph, test=develop
* fix comment for GradClipByGlobalNorm, test=develop
6 years ago
8777e8c1e9
fix Conv2DTranspose API, test=develop ( #21403 )
6 years ago
fbf9eca0d3
QAT Int8 document ( #21360 )
...
* update benchmark for int8v2, QAT1, QAT2 accuracy and performance
test=document_fix
* change according to reviews
test=develop test=document_fix
* improve some descriptions and some models
test=develop test=document_fix
* update models benchmark data
test=develop test=document_fix
* update int8v2 and qat2 performance
test=develop test=document_fix
6 years ago
a1a5adc9b8
Refine English Doc of cond API ( #21609 )
...
As the title
6 years ago
548efcd2e4
Fix unit tests to avoid check_grad checking failures ( #21554 )
...
* fix python API tests that do not need to inherit OpTest, test=develop
* fix fp16 cases that will only be enabled in GPU mode, test=develop
* remove TestSoftmaxFP16Op from test cases of softmax_mkldnn_op, test=develop
* fix tests so that the cases are only created in GPU mode, test=develop
6 years ago
4ad9b755c7
fix paddle compile errors under some python versions ( #21616 )
...
* fix compile error in some python version, test=develop
* remove redudant code, test=develop
6 years ago
be6a639655
Add CI for checking Input/Output/Attr of modified Ops ( #21522 )
...
* add shell scripts. test=develop
* rename test_pybind_inference to test_pybind_interface and print repeat process in check_op_desc.py. test=develop
* add approval RD. test=develop
6 years ago
835ea12cde
Control flow API: while_loop ( #21276 )
...
Add basic while_loop
6 years ago
4f81d1bd5f
Refine VarBase init function ( #21587 )
...
* refine init function, test=develop
* add tests, test=develop
* remove extern, which may cause symbol error in gcc-4.8, test=develop
6 years ago
56882ce432
change input data type and decrease max_relative_error value in test_check_grad for grop_nom_op test test=develop ( #21608 )
6 years ago
1dcf6a7212
Add Much Complex Test and Fix Bugs for Control Flow cond API ( #21532 )
...
Add tests to use dy/dx to make sure the gradient values calculated by the control flow backward is correct. Also fixed bugs detected by those tests.
Fix bugs:
1. Unlike sum_op, optimizer ops don't allow uninitialized input tensor. But in conditional_block_grad_op, since the conditional_block may not run, the output gradient tensor may be uninitialized, which will cause the optimizer op error. To fix it, we should let optimizer ops support uninitialized input like sum_op or assign the uninitialized gradient to 0 when the conditional_block_grad_op doesn't run. I found there are about 10+ optimizer ops. **To be simpler, I just assign output gradient of the conditional_block_grad_op to 0 in this PR**. But it can be further explored whether we can make optimizer ops like sum_op to support uninitialized input tensor because theoretically we can speed up without the assigning in conditional_block_grad_op.
2. Infer parameter shapes during append_backward. I didn't know that all our parameters are in global block. When op_desc is inferring shapes at the sub-block, it may not know the shape of gradients of parameters whose shape information is at global block. I fixed it by inferring shapes of gradients from forward var.
This PR also did some code clean up:
1. Print the var name when sgd_op catches shape error so that it is easier to debug
2. Fix a typo: dicta -> dict
6 years ago
c5aec2fe68
Paddlebox Related to Framework ( #21586 )
...
* Add a single_process_multi_thread transpiler.
* Add some UTs.
* Fix some API description.
6 years ago
9da7e6b4d4
add file check_op_desc.py and add interface to get default value. ( #21530 )
...
* add file check_op_desc.py and add interface to get default value. test=develop
* add test for c++ coverage rate. test=develop
* Correct typo. test=develop
6 years ago
2057df7ac0
add fluid.layers.gelu & doc ( #21515 )
...
Add a python interface for Gelu.
Add documentation for fluid.layers.gelu.
6 years ago
29c3844585
fix doc typo test=develop ( #21566 )
6 years ago
9ce0e29dc3
[MKL-DNN] Batch norm mkl-dnn NHWC support ( #21553 )
...
* - BAtch norm mkl-dnn NHWC
test=develop
- compilation fix
test=develop
- UT fix
- cosmetics
test=develop
- Fix to Batch Norm MKL-DNN NHWC UT
test=develop
Conflicts:
paddle/fluid/operators/batch_norm_op.h
* - Lint fixes
test=develop
6 years ago
657053f262
remove elementwise x_should_larger_than_y restriction;test=develop ( #21517 )
6 years ago
08483a6819
Add dygraph linear warm up decay ( #21186 )
...
* dygraph mode support linear lr warm up; test=develop
* add unitest for linear warmup; test=develop
* add input type check; test=develop
* fix type check assert error; test=develop
* change type error; test=develop
6 years ago
da75ac8b6c
bugfix: construct a DistributedStrategy instance if the passed one is None ( #21545 )
6 years ago
de46b15951
Unify the rank of prelu alpha to 4, corresponding to [N, C, H, W], except for the all mode
6 years ago
cdd46d7e02
Split VarBase from Python Variable for Dygraph ( #21359 )
...
* test=develop, fix docker with paddle nccl problem
* don't expose numerous Tensor.set(), test=develop
* fix condition, test=develop
* fix float16 bug, test=develop
* feed should be Tensor or np.array, not Variable or number, test=develop
* use forcecast to copy numpy slice to new array, test=develop
* remove float16-uint16 hacking, test=develop
* add variable method to varbase and refactor to_variable to support return varbase
* support kwargs in varbase constructor
* add VarBase constructor to support default python args
* refine varbase initial method
* reset branch
* fix ut for change VarBase error info to PaddleEnforce
* cherry is parameter change before
* overload isinstance to replace too many change of is_variable
* rm useless files
* rm useless code merged by git
* test=develop, fix some ut failed error
* test=develop, fix test_graph_wrapper
* add some tests, test=develop
* refine __getitem__, test=develop
* add tests, test=develop
* fix err_msg, test=develop
6 years ago
cdba41af4d
dygraph Embedding layer use lookuptable v2 ( #21209 )
...
* dygraph Embedding layer use lookuptable v2
test=develop
* fix test_nce
test=develop
6 years ago
4c9b3dafa7
fill_constant_batch_size_like OP precious problem fix ( #21337 )
...
* fix fill_constant_batch_size_like_op precious problem test=develop
6 years ago
b1da35261b
fix unit tests that do not need to inherit OpTest ( #21460 )
...
* fix PythonAPI test in Op unittest, test=develop
* fix unit tests that do not need to inherit OpTest, test=develop
6 years ago
768f9242e9
Fix dgc clip & rampup step, test=develop ( #21491 )
6 years ago
54382ce497
Add get_all_kernels api of registered data_type in pybind.cc ( #21499 )
...
* add _get_all_register_op_kernels api test=develop
* refine usage of check_op_register_type test=develop
* add import in core test=develop
6 years ago
e75ded08a0
fix the compatiable problem between PY2 and PY3 (issue#20749) ( #20942 )
...
* fix the compatiable problem between PY2 and PY3.
* add ut, test=develop
* add proxy, test=develop
* download dataset before test, test=develop
6 years ago
a3535812f6
add _use_system_allocator to some op tests, test=develop ( #21504 )
6 years ago
18a5d30754
[MKL-DNN] Conv2d and Conv2d transpose MKL-DNN NHWC support ( #21466 )
6 years ago
2445fef386
Fix density sample ( #21506 )
6 years ago
6ebf0f47b8
support SelectedRows in dygraph, test=develop ( #21078 )
...
* support SelectedRows in dygraph, test=develop
* fix bug of _grad_ivar interface, test=develop
* add optest for support seletedrows, test=develop
* fix bug for gradient_accumulator in GPU mode, test=develop
* fix error when Selectedrows addto LodTensor in sorted_gradient mdoe in dygraph, test=develop
* refine and simplify gradient accumulator code, test=develop
* add optest, test=develop
* add optest and simplify code, test=develop
* fix bug for test_imperative_selected_rows, test=develop
* add optest for Coverage, test=develop
* fix gradient interface and simplify code, test=develop
* update api for gradient, test=develop
* fix ShareDim's bug in DygraphExecutionContext class, test=develop
* add optest, test=develop
6 years ago
0bc8bdf724
set dim[0] to -1 if dim[0] < 0 during compiling for c_allgather op ( #21402 )
...
* set dim[0] to -1 if dim[0] < 0 and remove assertion to runtime, test=develop
* modify ENFORCE message, test=develop
* add validation for x.shape[0] > 0, test=develop
* add ut, test=develop
6 years ago
4bf115b42d
Fix AdamOptimizer and Scale sample code Bug ( #21478 )
...
* fix adam sample code bug test=document_fix
* fix sample code bug in scale test=document_fix
6 years ago
94bef03539
Revert "Add masked select api ( #21172 )" ( #21456 )
...
This reverts commit 007c997572
6 years ago
3706ea67f8
fix sample code in density prior box
6 years ago
87ab93af01
fix adam fp64, test=develop ( #21423 )
6 years ago
beec87b911
fix bug in example codes of API case and switch_case. test=develop,test=document_fix ( #21477 )
6 years ago
7e68bc896b
refactor AUC OP and add its CUDA Kernel ( #21336 )
...
* refactor AUC OP and add its CUDA Kernel
* the layout of global auc doesn't change
6 years ago
1f57ac1241
delete concat in AddQuantDequantPass, test=develop ( #21454 )
6 years ago
2a54c359f0
add fraction of cpu memory to use, test=develop ( #21453 )
6 years ago
101240d2c1
fix PythonAPI test in Op unittest, test=develop ( #21462 )
...
There are PythonAPI tests in Op's unittest which don't need to inherit OpTest class.
6 years ago
dbbe6e9cb6
fix the device supported of the op unique and unique_with_counts. ( #21395 )
...
* fix the device supported of the op unique and unique_with_counts.
test=develop
test=document_fix
* Fix the precision of test in the op of unique and unique_with_counts.
test=develop
test=document_fix
6 years ago
32959e031e
Add English Document for cond API ( #21452 )
...
Add English doc for cond
6 years ago
3df13ab40c
fix PythonAPI test in Op unittest, test=develop ( #21455 )
...
There are PythonAPI tests in Op's unittest which don't need to inherit OpTest class.
6 years ago
5e813b53c5
nhwc optimization for batchnorm ( #21090 )
6 years ago
e0c9d856fb
add unused input vars check for OpWithKernel, test=develop ( #21169 )
...
* add unused input vars check for OpWithKernel, test=develop
* remove unused vars in some ops, test=develop
* fix batch_norm, test=develop
* add white list, test=develop
* add CI check for white list, test=develop
* :ove white list to c++, test=develop
* solve failure of CI, test=develop
* add unittest for unused_var_check, test=develop
* refine code, enable check in operator_test, test=develop
* skip mkldnn, test=develop
* extend white list, test=develop
* refine condition of mkldnn, test=develop
* fix paddle_build, test=develop
* follow comments, test=develop
* fix GetExpectedKernelType
* add wiki ref to err_msg, test=develop
* follow comment, test=develop
6 years ago
664f958a02
Fix optimizer op infershape failed in dygraph multi-cards mode ( #21374 )
...
* add param & grad shape check for sgd op
* add _reshape_inplece interface for dygraph parallel
* refine unittest based paddle/models scripts, test=develop
* add unittest for parallel grad fuse, test=develop
6 years ago
630be31952
Fix Cond Bug for Nested Control Flow ( #21340 )
...
* Commit before merging develop
test=develop
* Backup after working with Huihuang logs
* Commit before deleting Huihuang debug loggings
* Commit before debug
test=develop
* Fix bug commit
test=develop
* Backup of fixing bugs
test=develop
* Clean up code
test=develop
* Fix a bug in sum_op
test=develop
6 years ago
cd43c4440e
[MKL-DNN] LRN and Pool2d (FWD) NHWC support ( #21375 )
6 years ago
ca879e5a77
fix skip_op bug ( #21418 )
...
test=develop
6 years ago
b16274556a
Add dscending for argsort ( #21400 )
...
* Add ascending for argsort
* Refine api doc description.
* Refine descending description
* Add int32 logic to speedup when data is small size.
* Remove int32 opt as not support in python
6 years ago
ac8546701d
Add dygraph execution context ( #20157 )
...
* add_dygraph_execution_context
* add dygraph infershape context and execution context; test=develop
* fix imperative bug; test=develop
* remove inputs outputs interface from execution context,
because it have same function with inputNames;
test=develop
* remove tracer_test ctest; test=develop
* fix split op bug; test=develop
* fix unitests bug; test=develop
* fix distribute test bug; test=develop
* fix ngraph compile bug; test=develop
* fix grad maker bug; test=develop
* fix load op bugs; test=develop
* fix operator.cc construct bug; test=develop
* remove useless name find in operator; test=develop
* add tracer_test; test=develop
* fix concat, split bug; test=develop
* remove tracer_test unitest; test=develop
* fix attribute check bug; test=develop
* add test code to fix converage; test=develop
* remove useless code, change check backward input in engin; test=develop
* unlock var type infer shape;test=develop
* add ShareAllLoD api; test=develop
* add dygraph infershape context unitest; test=develop
* remove increase and decrease lod in dygraph; test=develop
* addd override; test=develop
* fix increase descrease lod; test=develop
* fix paddle_enforce; test=develop
* disable lod op dygraph check; test=develop
* fix paddle enforce error; test=develop
* add comment for op_registry and OperatorBase; test=develop
* optimize the comment of op_registry; test=develop
* fix format of comment; test=develop
* fix format of comment; test=develop
* optimize the format of comment; test=develop
* optimize the format of the comment; test=develop
* optimize comment of op_registry; test=develop
6 years ago
a6b089c614
add macro to ban windows ( #21422 )
...
remove nccl related code in windows
6 years ago
ebfb720a63
add Adam beta1/beta2 support Variable ( #21234 )
...
* add Adam beta1/beta2 support Variable. test=develop
6 years ago
09696d5df8
Use system allocator in OpTest ( #21335 )
...
* use system allocator in unittests, test=develop
* fix op bugs, test=develop
* fix tensor copy bug when src and dst are the same, test=develop
6 years ago
007c997572
Add masked select api ( #21172 )
6 years ago
67c836fb5c
batch_norm momentum support variable ( #21246 )
...
* batch_norm momentum support variable. test=develop
* fix format. test=develop
* add batch_norm momentum variable example. test=develop
* move MomentumTensor to training branch. test=develop
* split example. test=develop
* fix doc. test=develop
* fix PADDLE_ENFORCE ci. test=develop
* fix format. test=develop
6 years ago
c0aa13672e
Fp32 vs int8 qat C++ performance ( #21244 )
...
* add ut for comparing FP32 and QAT INT8
* add save qat transformed model python script
test=develop
* updated
* added missing file
* add "with_label"
test=develop
* performance benchmark as unit test
test=develop
* change names of unnecessary thing
* Change CMakeList.txt for model downloading and UT
test=develop
* change names of functions and params for more readable code
test=develop
* Change PADDLE_ENFORCE messages
test=develop
* fix indent problems
test=develop
* indent problems
test=develop
6 years ago
f1178e9d79
fix fleet save bug ( #21362 )
...
* fix fleet save bug of save_infernece_model
* test=develop
6 years ago
Liufang Sang
1840c1652c
add config file to avoid load checkpoint test=develop ( #21373 )
6 years ago
b97fc16d21
fix lod_reset bug, test=develop ( #21392 )
6 years ago
47a82e38e3
Support data_norm gpu kernel ( #21325 )
...
* support data_norm_op run in CUDA
* add two parameters sync_stats & summary_decay_rate
* add UT
6 years ago
d5ff79e55e
Support numpy bridge (enabled by default in dygraph mode) ( #20983 )
...
* add numpy bridge
* fix template compile
* add unittest, add default
test=develop
* fix unittest
test=develop
* fix unittest
test=develop
* zero_copy=True for to_variable,
test=develop
* bug fix
test=develop
* disable deprecated NumPy API
test=develop
* use better design of NumpyAllocator
test=develop
* fix Py_None check
test=develop
* reset c++ tracer when jump out dygraph guard
test=develop
* refine PADDLE_ENFORCE_xx format
test=develop
* bug fix of tracer switch
test=develop
* update decref
test=develop
6 years ago
5d7d548275
INT8 Fully-connected ( #17641 )
...
* Implement Int8 FC
* Integrate FC into INT8v2
test=develop
* int8 FC: transpose weights before computing scales
test=develop
* Add support for activation_type string in FC
test=develop
* Disable MKL-DNN's FC in VGG16 and 19
test=develop
* Disable FC quantization when mkldnn FC is disabled
test=develop
* Solve PADDLE_ENFORCES in FC int8
* Fix Paddle enforces and remove const cast
test=develop
* Fix style changes
test=develop
* Fix quantizer_tester test and add fc quantization
test=develop
* Fix FC test fail on CUDA
* Remove unnecessary log from quantize placement pass
test=develop
* Add Thread ID to FC hash key
test=develop
* Add comments to MKL-DNN FC Kernel
test=develop
* Refactor quantizer
test=develop
* Fix linter issues
test=develop
* Fix crash in slim googlenet
test=develop
* Fix PADDLE_ENFORCE messages
test=develop
6 years ago
07e6a94268
paddleslim quantization skip pattern support list of string ( #21141 )
6 years ago
be2e3e67d9
Fix some typos in AMP. ( #21354 )
...
* fix some typos in AMP. test=develop
* delete useless codes. test=develop
6 years ago
zhongpu
songyouwei
Leo Chen
Chen Weihang
Zhang Ting
Zeng Jinle
Double_V
mapingshuo
gongweibao
juncaipeng
zhupengyang
hong
Chengmo
Yibing Liu
liym27
Feiyu Chan
WangXi
Jacek Czaja
Huihuang Zheng
Qinghe JING
silingtong123
chengjuntao
123malin
Adam
ShenLiang
Leo Zhao
Liufang Sang
Michał Gallus
huzhiqiang
FDInSky
ceci3
dyning
Pei Yang
tianshuo78520a
Aurelius84
Yang Zhang
Kaipeng Deng
Li Fuchen
zhoushiyu
pkpk
lijianshe02
zhouwei25
hutuxian
GaoWei8
zhupengyang
Aurelius84
lidanqing
guofei
Guo Sheng
Galden
tangwei12
Wojciech Uss
zhaoyuchen2018
lilong12
joanna.wozna.intel
baojun
xujiaqi01
wangchaochaohu
danleifeng
ruri
wawltor
Jie Fang
itminner
Zhen Wang