| .. |
|
amp
|
[ROCM] update fluid collective op for rocm, test=develop (#31075)
|
4 years ago |
|
benchmark
|
[ROCM] update fluid collective op for rocm, test=develop (#31075)
|
4 years ago |
|
collective
|
c_gen_nccl_id add SocketServer to persit server (#31589)
|
4 years ago |
|
controlflow
|
[ROCM] update fluid operators for rocm (part1), test=develop (#31077)
|
4 years ago |
|
detail
|
[ROCM] update fluid collective op for rocm, test=develop (#31075)
|
4 years ago |
|
detection
|
Polish two error messages (#31852)
|
4 years ago |
|
distributed
|
[ROCM] update fluid operators for rocm (part1), test=develop (#31077)
|
4 years ago |
|
distributed_ops
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
elementwise
|
[oneDNN] Added Elementwise Mul grad fp32/bf16 (#31647)
|
4 years ago |
|
fused
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
jit
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
lite
|
[ROCM] update fluid operators for rocm (part3), test=develop (#31213)
|
4 years ago |
|
math
|
[oneDNN] lookup_table op with support for BF16 data type. (#31558)
|
4 years ago |
|
metrics
|
[ROCM] update fluid operators for rocm (part1), test=develop (#31077)
|
4 years ago |
|
mkldnn
|
fix cache key in concat oneDNN kernel (#31820)
|
4 years ago |
|
nccl
|
[ROCM] update fluid framework for rocm (part6), test=develop (#31015)
|
4 years ago |
|
optimizers
|
lamb_op_xpu;test=kunlun (#31012)
|
4 years ago |
|
pscore
|
fix compilation errors for missing brpc header files, test=develop (#31325)
|
4 years ago |
|
reader
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
reduce_ops
|
[ROCM] fix reduce_sum nan in ROCM platform, test=develop (#31780)
|
4 years ago |
|
sequence_ops
|
[ROCM] fix dropout and remove hipcub, test=develop (#31455)
|
4 years ago |
|
tensorrt
|
[Paddle-TRT] Fix engine key in trt int8 calibration (#31513)
|
4 years ago |
|
CMakeLists.txt
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
abs_op.cc
|
make abs op support complex types (#30375)
|
4 years ago |
|
abs_op.cu
|
make abs op support complex types (#30375)
|
4 years ago |
|
abs_op.h
|
make abs op support complex types (#30375)
|
4 years ago |
|
activation_cudnn.cu.cc
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
activation_cudnn_op.cu.cc
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
activation_op.cc
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
activation_op.cu
|
relu forward and backward with vectortype (#31869)
|
4 years ago |
|
activation_op.h
|
fix ELU output for nan, test=develop (#31132)
|
4 years ago |
|
activation_op_xpu.cc
|
update activation op on kunlun (#29577)
|
4 years ago |
|
add_position_encoding_op.cc
|
Polish no onwer ops error message (#27448)
|
4 years ago |
|
add_position_encoding_op.h
|
OP(warpctc, add_position_encoding, scaled_dot_product_attention) error message enhancement (#24261)
|
5 years ago |
|
addmm_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
addmm_op.cu
|
test=develop, add addmm op (#23384)
|
5 years ago |
|
addmm_op.h
|
test=develop, add addmm op (#23384)
|
5 years ago |
|
affine_channel_op.cc
|
API/OP (affine_channel, group_norm, layer_norm, random_crop, unpool, … (#24118)
|
5 years ago |
|
affine_channel_op.cu
|
Bugfix rocm (#31490)
|
4 years ago |
|
affine_channel_op_xpu.cc
|
support roi_align & affine_channel for kunlun (#29561)
|
4 years ago |
|
affine_grid_cudnn_op.cu.cc
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
affine_grid_op.cc
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
affine_grid_op.cu
|
Fix cuda kernel of affine grid (#27003)
|
5 years ago |
|
affine_grid_op.h
|
【2.0 API】Enhance affine grid operator (#26385)
|
5 years ago |
|
allclose_op.cc
|
Fix the accuracy problem of allclose op when using float64 data type in static mode. (#29890)
|
4 years ago |
|
allclose_op.cu
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
allclose_op.h
|
Fix the accuracy problem of allclose op when using float64 data type in static mode. (#29890)
|
4 years ago |
|
arg_max_op.cc
|
add the op version check for the elementwise ops, test=op_version (#30010)
|
4 years ago |
|
arg_max_op.cu
|
add cub impl for arg max, min (#25941)
|
5 years ago |
|
arg_min_max_op_base.cu.h
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
arg_min_max_op_base.h
|
optimize compilation time of argmin/argmax op (#29595)
|
4 years ago |
|
arg_min_op.cc
|
add the op version check for the elementwise ops, test=op_version (#30010)
|
4 years ago |
|
arg_min_op.cu
|
add cub impl for arg max, min (#25941)
|
5 years ago |
|
argsort_op.cc
|
rename inplace/no_need_buffer inferer, part2, test=develop (#24733)
|
5 years ago |
|
argsort_op.cu
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
argsort_op.h
|
Correct CPU gradients of the argsort op (#22739)
|
5 years ago |
|
array_operator.h
|
fix bugs in transformer predict in xpu place (#30730)
|
4 years ago |
|
array_to_lod_tensor_op.cc
|
[ROCM] update fluid operators for rocm (part5), test=develop (#31258)
|
4 years ago |
|
ascend_trigger_op.cc
|
Ascend Framework Part1: OP & Wrapper (#30281)
|
4 years ago |
|
ascend_trigger_op.h
|
Ascend Framework Part1: OP & Wrapper (#30281)
|
4 years ago |
|
assert_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
assign_op.cc
|
[ROCM] update fluid operators for rocm (part5), test=develop (#31258)
|
4 years ago |
|
assign_op.h
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
assign_op_test.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
assign_op_xpu.cc
|
add cast/concat/assign xpu op (#27911)
|
4 years ago |
|
assign_value_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
assign_value_op.cu.cc
|
Add the support of bool list for assign_value op (#23774)
|
5 years ago |
|
assign_value_op.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
attention_lstm_op.cc
|
modify error message based on comments (#30189)
|
4 years ago |
|
attention_lstm_op.h
|
…
|
|
|
average_accumulates_op.cc
|
c++ API ( average_accumulates, tensor_array_to_tensor and average_accumulates) error message enhance. test=develop (#23631)
|
5 years ago |
|
average_accumulates_op.cu
|
Add macro BOOST_GET to enrich the error information of boost :: get (#24175)
|
5 years ago |
|
average_accumulates_op.h
|
optimize unity build (#30195)
|
4 years ago |
|
batch_fc_op.cc
|
rename inplace/no_need_buffer inferer, part2, test=develop (#24733)
|
5 years ago |
|
batch_fc_op.cu
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
batch_fc_op.h
|
Add batch_fc op in contrib (#24017)
|
5 years ago |
|
batch_norm_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
batch_norm_op.cu
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
batch_norm_op.h
|
support channel last in BatchNorm*d
|
4 years ago |
|
batch_norm_op_xpu.cc
|
optimize batch_norm & pool op for kunlun (#30490)
|
4 years ago |
|
batch_size_like.h
|
refine the error message for bath size like OP (#27446)
|
4 years ago |
|
bce_loss_op.cc
|
change to use bce_loss op, add shape check for bce_loss
|
5 years ago |
|
bce_loss_op.cu
|
[ROCM] update fluid operators for rocm (part6), test=develop (#31301)
|
4 years ago |
|
bce_loss_op.h
|
change to use bce_loss op, add shape check for bce_loss
|
5 years ago |
|
beam_search_decode_op.cc
|
Optimize the error message for OP (#27617)
|
4 years ago |
|
beam_search_decode_op.h
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
beam_search_decode_op_test.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
beam_search_op.cc
|
Fix beam_search InferShape (#25169)
|
5 years ago |
|
beam_search_op.cu.cc
|
…
|
|
|
beam_search_op.h
|
API/OP(sequence_first_step, sequence_last_step, sequence_mask, beam_search, beam_search_decode) error message enhancement (#24590)
|
5 years ago |
|
bernoulli_op.cc
|
Refine paddle.manual_seed (#26496)
|
5 years ago |
|
bernoulli_op.cu
|
use cuda generator in bernoulli cuda kernel (#30199)
|
4 years ago |
|
bernoulli_op.h
|
Refine bernoulli and unsqueeze op (#26842)
|
5 years ago |
|
bilateral_slice_op.cc
|
Fix bilateral inference shape bug (#26822)
|
4 years ago |
|
bilateral_slice_op.cu
|
Enable bilateral_slice unittest on windows platform (#29896)
|
4 years ago |
|
bilateral_slice_op.h
|
Add bilateral_slice op (#25401)
|
5 years ago |
|
bilinear_tensor_product_op.cc
|
modify error message based on comments (#30189)
|
4 years ago |
|
bilinear_tensor_product_op.cu
|
…
|
|
|
bilinear_tensor_product_op.h
|
fix typo words (#22653)
|
5 years ago |
|
bmm_op.cc
|
create bmm op and move several api from fluid.layers to tensor (#23457)
|
5 years ago |
|
bmm_op.cu
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
bmm_op.h
|
create bmm op and move several api from fluid.layers to tensor (#23457)
|
5 years ago |
|
bpr_loss_op.cc
|
enhance cvm bpr_loss adam adagrad adamax ftrl error message, test=develop (#24452)
|
5 years ago |
|
bpr_loss_op.h
|
enhance cvm bpr_loss adam adagrad adamax ftrl error message, test=develop (#24452)
|
5 years ago |
|
cast_op.cc
|
[oneDNN] Initial bf16 amp integration (#31093)
|
4 years ago |
|
cast_op.cu
|
add VecCastCUDAKernel (#30296)
|
4 years ago |
|
cast_op.h
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
cast_op_xpu.cc
|
support some shape for matmul and cast in xpu place (#29900)
|
4 years ago |
|
center_loss_op.cc
|
API/OP (center_loss, fluid.one_hot, prroi_pool, roi_pool, ctc_greed_decoder) error message enhancement (#23794)
|
5 years ago |
|
center_loss_op.cu
|
Error message opt, test=develop (#27467)
|
4 years ago |
|
center_loss_op.h
|
Add center Loss Op Support (#18681)
|
6 years ago |
|
cholesky_op.cc
|
Add cholesky_op (#23543)
|
5 years ago |
|
cholesky_op.cu
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
cholesky_op.h
|
add error message for cholesky (#26444)
|
5 years ago |
|
chunk_eval_op.cc
|
API(dynamic_gru, chunk_eval, BeamSearchDecoder) error message enhancement (#24513)
|
5 years ago |
|
chunk_eval_op.h
|
Optimize the error message for OP (#27617)
|
4 years ago |
|
clip_by_norm_op.cc
|
…
|
|
|
clip_by_norm_op.cu
|
…
|
|
|
clip_by_norm_op.h
|
Fix the nan bug when passing all zero values into clip_by_norm_op. (#30777)
|
4 years ago |
|
clip_by_norm_op_xpu.cc
|
add clip_by_norm on kunlun, *test=kunlun (#30862)
|
4 years ago |
|
clip_op.cc
|
Add clip double grad (#29590)
|
4 years ago |
|
clip_op.cu
|
add clamp api, test=develop (#23273)
|
5 years ago |
|
clip_op.h
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
coalesce_tensor_op.cc
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
common_infer_shape_functions.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
common_infer_shape_functions.h
|
Add broadcast_shape api (#28257)
|
4 years ago |
|
concat_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
concat_op.cu.cc
|
Add support for tuple of concat Op test=develop (#25800)
|
5 years ago |
|
concat_op.h
|
fix concat dimension (#25606)
|
5 years ago |
|
concat_op_xpu.cc
|
fix bugs in transformer predict in xpu place (#30730)
|
4 years ago |
|
conj_op.cc
|
add conj op for complex types (#29527)
|
4 years ago |
|
conj_op.cu
|
add conj op for complex types (#29527)
|
4 years ago |
|
conj_op.h
|
complex gradient matmul (#29966)
|
4 years ago |
|
conv_cudnn_helper.h
|
enable exhaustive_search for forward and backward algos when dtype is float16 (#30959)
|
4 years ago |
|
conv_cudnn_op.cu
|
[ROCM] fix conv2d and conv3d op, test=develop (#31553)
|
4 years ago |
|
conv_cudnn_op_cache.h
|
Delete cudnn6 code (#31835)
|
4 years ago |
|
conv_miopen_helper.h
|
[ROCM] fix conv2d and conv3d op, test=develop (#31553)
|
4 years ago |
|
conv_op.cc
|
update conv2d, test=develop (#31480)
|
4 years ago |
|
conv_op.cu.cc
|
…
|
|
|
conv_op.h
|
support channel last in BatchNorm*d
|
4 years ago |
|
conv_op_xpu.cc
|
add squeeze_op/unsqueeze_op on kunlun;fix conv op and parallel executor;optimize lookup_table op (#31056)
|
4 years ago |
|
conv_shift_op.cc
|
Remove extraneous comma in error messages (#24478)
|
5 years ago |
|
conv_shift_op.cu
|
…
|
|
|
conv_shift_op.h
|
…
|
|
|
conv_transpose_cudnn_op.cu
|
[ROCM] fix test_conv2d_transpose_op (#31749)
|
4 years ago |
|
conv_transpose_op.cc
|
[ROCM] update fluid operators for rocm (part4), test=develop (#31225)
|
4 years ago |
|
conv_transpose_op.cu
|
Add double grad for conv_transpose (#29706)
|
4 years ago |
|
conv_transpose_op.h
|
fix bug of DepthwiseConvTransposeGradKernel (#31762)
|
4 years ago |
|
correlation_op.cc
|
Add correlation api to contrib (#27015)
|
4 years ago |
|
correlation_op.cu
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
cos_sim_op.cc
|
Fix ops doc for some ops
|
4 years ago |
|
cos_sim_op.cu
|
…
|
|
|
cos_sim_op.h
|
…
|
|
|
crf_decoding_op.cc
|
Format error message for ops (#24482)
|
5 years ago |
|
crf_decoding_op.h
|
Format error message for ops (#24482)
|
5 years ago |
|
crop_op.cc
|
rename inplace/no_need_buffer inferer, part4, test=develop (#24781)
|
5 years ago |
|
crop_op.cu
|
fix document of 11 APIs (#20278)
|
5 years ago |
|
crop_op.h
|
Improving error reporting messages for ops (#24438)
|
5 years ago |
|
crop_tensor_op.cc
|
Improving error reporting messages for ops (#24438)
|
5 years ago |
|
crop_tensor_op.cu
|
All elements in attr(shape) of crop_tensor can be -1 and int32/64 kernel registered (#20756)
|
5 years ago |
|
crop_tensor_op.h
|
Fix the bug that Input(Offsets) and attr(offsets) cannot be set at the same time. (#24975)
|
5 years ago |
|
cross_entropy_op.cc
|
Fix bug: Don't check dims if contain_unknown_dim of cross_entropy_grad_op in compile time (#25221)
|
5 years ago |
|
cross_entropy_op.cu
|
…
|
|
|
cross_entropy_op.h
|
Enhance error message of cross_entropy_op, sigmoid_cross_entropy_with_logits_op (#24485)
|
5 years ago |
|
cross_op.cc
|
add new api for Paddle2.0: nonzero, index_selct, roll, cross (#23176)
|
5 years ago |
|
cross_op.cu
|
add new api for Paddle2.0: nonzero, index_selct, roll, cross (#23176)
|
5 years ago |
|
cross_op.h
|
add new api for Paddle2.0: nonzero, index_selct, roll, cross (#23176)
|
5 years ago |
|
ctc_align_op.cc
|
API/OP (center_loss, fluid.one_hot, prroi_pool, roi_pool, ctc_greed_decoder) error message enhancement (#23794)
|
5 years ago |
|
ctc_align_op.cu
|
Error message opt, test=develop (#27467)
|
4 years ago |
|
ctc_align_op.h
|
Error message opt, test=develop (#27467)
|
4 years ago |
|
cudnn_lstm_cache.h
|
Delete cudnn6 code (#31835)
|
4 years ago |
|
cudnn_lstm_op.cc
|
add REGISTER_OP_VERSION for LSTM (#30038)
|
4 years ago |
|
cudnn_lstm_op.cu.cc
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
cudnn_rnn_cache.h
|
Delete cudnn6 code (#31835)
|
4 years ago |
|
cum_op.h
|
fix cumsum op for API 2.0, optimize performance
|
5 years ago |
|
cumsum_op.cc
|
register cumsum Op version for compatible Op upgrades (#26734)
|
5 years ago |
|
cumsum_op.cu
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
cvm_op.cc
|
enhance error message, test=develop (#30220)
|
4 years ago |
|
cvm_op.cu
|
Fix index overflow bug of the CUDA kernel loop increment (#25435)
|
5 years ago |
|
cvm_op.h
|
mod cvm test=develop (#25146)
|
5 years ago |
|
data_norm_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
data_norm_op.cu
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
data_norm_op.h
|
…
|
|
|
deformable_conv_filter.cu.h
|
add deformable conv v1 op and cpu version of deformable conv v2 (#18500)
|
5 years ago |
|
deformable_conv_func.h
|
add deformable conv v1 op and cpu version of deformable conv v2 (#18500)
|
5 years ago |
|
deformable_conv_op.cc
|
Refine error message, test=develop (#23823)
|
5 years ago |
|
deformable_conv_op.cu
|
add deformable conv v1 op and cpu version of deformable conv v2 (#18500)
|
5 years ago |
|
deformable_conv_op.h
|
Fix compling warning in deformable conv. (#20036)
|
5 years ago |
|
deformable_conv_op_xpu.cc
|
fix expand/uniform_random && concat/transpose to new api on xpu (#29280)
|
4 years ago |
|
deformable_conv_v1_op.cc
|
Refine error message, test=develop (#23823)
|
5 years ago |
|
deformable_conv_v1_op.cu
|
add deformable conv v1 op and cpu version of deformable conv v2 (#18500)
|
5 years ago |
|
deformable_conv_v1_op.h
|
Fix compling warning in deformable conv. (#20036)
|
5 years ago |
|
deformable_psroi_pooling_op.cc
|
OP(retinanet_detection_output, retinanet_target_assign, sigmoid_focal_loss, deformable_roi_pooling) error message enhancement. test=develop (#23726)
|
5 years ago |
|
deformable_psroi_pooling_op.cu
|
Fix index overflow bug of the CUDA kernel loop increment (#25435)
|
5 years ago |
|
deformable_psroi_pooling_op.h
|
OP(retinanet_detection_output, retinanet_target_assign, sigmoid_focal_loss, deformable_roi_pooling) error message enhancement. test=develop (#23726)
|
5 years ago |
|
delete_var_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
dequantize_abs_max_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
dequantize_abs_max_op.cu
|
add dequantize_abs_max op and modify lookup_table op (#20899)
|
5 years ago |
|
dequantize_abs_max_op.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
dequantize_log_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
dequantize_log_op.cu
|
remove pow to speed up in dequantize_log op (#24607)
|
5 years ago |
|
dequantize_log_op.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
dequantize_op.cc
|
operator checkpoints for new attributes. (#29832)
|
4 years ago |
|
dequantize_op.h
|
…
|
|
|
dequeue_op.cc
|
add queue_generator_op, dequeue_op, enqueue_op and ut (#24481)
|
5 years ago |
|
detection_map_op.cc
|
optimize the error meesage for detetion_map_op
|
4 years ago |
|
detection_map_op.h
|
optimize the error meesage for detetion_map_op
|
4 years ago |
|
dgc_clip_by_norm_op.cc
|
Optimize error message, include dgc, nccl, size op (#24456)
|
5 years ago |
|
dgc_clip_by_norm_op.cu
|
…
|
|
|
dgc_clip_by_norm_op.h
|
…
|
|
|
dgc_op.cc
|
Optimize error message, include dgc, nccl, size op (#24456)
|
5 years ago |
|
dgc_op.cu
|
…
|
|
|
dgc_op.h
|
Replace all errors thrown by LOG(FATAL) with PADDLE_THROW (#24759)
|
5 years ago |
|
diag_embed_op.cc
|
add diag_embed op (#23385)
|
5 years ago |
|
diag_embed_op.cu
|
add diag_embed op (#23385)
|
5 years ago |
|
diag_embed_op.h
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
diag_op.cc
|
API(argsort, argmax, argmin, cast, diag) error message enhancement
|
5 years ago |
|
diag_op.cu
|
…
|
|
|
diag_op.h
|
…
|
|
|
diag_v2_op.cc
|
Fix diag OP bug on Windows Python3.8
|
4 years ago |
|
diag_v2_op.cu
|
Fix bug: The calculation result of Diag_v2 Op under large size input is wrong (#27447)
|
4 years ago |
|
diag_v2_op.h
|
add paddle.tensor.linalg.diag API, diag_v2 OP and CUDA kernel
|
5 years ago |
|
dist_op.cc
|
add dist op (#23503)
|
5 years ago |
|
dist_op.cu
|
[ROCM] fix test_dist_op ci test, test=develop (#31468)
|
4 years ago |
|
dist_op.h
|
use eval to improve performance, test=develop (#25459)
|
4 years ago |
|
dot_op.cc
|
complex gradient matmul (#29966)
|
4 years ago |
|
dot_op.cu
|
complex gradient matmul (#29966)
|
4 years ago |
|
dot_op.h
|
[ROCM] fix test_matmul_v2_op (#31802)
|
4 years ago |
|
dropout_op.cc
|
API/OP error message enhancement (#23691)
|
5 years ago |
|
dropout_op.cu
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
dropout_op.h
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
dropout_op_test.cc
|
…
|
|
|
dropout_op_xpu.cc
|
Polish kunlun error (#27974)
|
4 years ago |
|
edit_distance_op.cc
|
Format error message for ops (#24482)
|
5 years ago |
|
edit_distance_op.cu
|
Optimize the error message for OP (#27617)
|
4 years ago |
|
edit_distance_op.h
|
Optimize the error message for OP (#27617)
|
4 years ago |
|
empty_op.cc
|
add empty_like op (python, and unit test), use c++ implementation of empty op, (#27287)
|
4 years ago |
|
empty_op.cu.cc
|
add empty op (c++, python, unit test) (#26659)
|
4 years ago |
|
empty_op.h
|
add empty op (c++, python, unit test) (#26659)
|
4 years ago |
|
enqueue_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
erf_op.cc
|
add approximation for gelu, test=develop (#22961)
|
5 years ago |
|
erf_op.cu
|
add erf op (#21785)
|
5 years ago |
|
erf_op.h
|
add erf op (#21785)
|
5 years ago |
|
expand_as_op.cc
|
Optimize the error message for OP (#27617)
|
4 years ago |
|
expand_as_op.cu
|
add register op_data_type of pad/expand_as et.al (#21718)
|
5 years ago |
|
expand_as_op.h
|
Optimize the error message for OP (#27617)
|
4 years ago |
|
expand_as_v2_op.cc
|
update expand as op to use the shape of the target tensor instead of the target tensor itself. (#29020)
|
4 years ago |
|
expand_as_v2_op.cu
|
Improve expand as (#26290)
|
5 years ago |
|
expand_as_v2_op.h
|
update expand as op to use the shape of the target tensor instead of the target tensor itself. (#29020)
|
4 years ago |
|
expand_op.cc
|
add double grad for expand (#27183)
|
4 years ago |
|
expand_op.cu
|
register fp16 kernel for some ops (#22650) (#22696)
|
5 years ago |
|
expand_op.h
|
fix expand/uniform_random && concat/transpose to new api on xpu (#29280)
|
4 years ago |
|
expand_v2_op.cc
|
fix the bug in expand_v2 op (#30984)
|
4 years ago |
|
expand_v2_op.cu
|
[API 2.0] adaptive expand op to use shape instead of expand_times (#26206)
|
5 years ago |
|
expand_v2_op.h
|
[API 2.0] adaptive expand op to use shape instead of expand_times (#26206)
|
5 years ago |
|
eye_op.cc
|
support Baidu Kunlun AI Accelerator (#25959)
|
5 years ago |
|
eye_op.cu
|
add eye op, kernel and unitest test=develop (#18980)
|
6 years ago |
|
eye_op.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
fake_dequantize_op.cc
|
add op version for fake_quant and fake_dequant ops, test=op_version (#29923)
|
4 years ago |
|
fake_dequantize_op.cu
|
Fix fake_quant error when cout > 1024, test=develop (#28603)
|
4 years ago |
|
fake_dequantize_op.h
|
[Quantization] Conv2d_transpose and mul support channnelwise quantization (#25639)
|
5 years ago |
|
fake_quantize_op.cc
|
[dygraph qat] Use layer to calculate output scale (#31861)
|
4 years ago |
|
fake_quantize_op.cu
|
[dygraph qat] Use layer to calculate output scale (#31861)
|
4 years ago |
|
fake_quantize_op.h
|
[dygraph qat] Use layer to calculate output scale (#31861)
|
4 years ago |
|
fc_op.cc
|
Polish the error message of fc, fused_fc_elementwise_layernorm and fused_embedding_seq_pool. (#27692)
|
4 years ago |
|
fc_op.cu.cc
|
Implement the GPU kernel of fc operator (#19687)
|
6 years ago |
|
fc_op.h
|
Polish the error message of fc, fused_fc_elementwise_layernorm and fused_embedding_seq_pool. (#27692)
|
4 years ago |
|
fill_any_like_op.cc
|
Add macro BOOST_GET to enrich the error information of boost :: get (#24175)
|
5 years ago |
|
fill_any_like_op.cu
|
Add the zeros, ones, ones_like, zeros_like for api 2.0, test=develop (#23471)
|
5 years ago |
|
fill_any_like_op.h
|
Add the error raise for some operators, add some test cases
|
5 years ago |
|
fill_constant_batch_size_like_op.cc
|
rename inplace/no_need_buffer inferer, part2, test=develop (#24733)
|
5 years ago |
|
fill_constant_batch_size_like_op.cu.cc
|
Add seq2seq api related code (#19820)
|
5 years ago |
|
fill_constant_batch_size_like_op.h
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
fill_constant_op.cc
|
Recompute Offload (#30233)
|
4 years ago |
|
fill_constant_op.cu.cc
|
Add complex dtype op (add) test example (#29603)
|
4 years ago |
|
fill_constant_op.h
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
fill_constant_op_xpu.cc
|
add conj op for complex types (#29527)
|
4 years ago |
|
fill_op.cc
|
Add macro BOOST_GET to enrich the error information of boost :: get (#24175)
|
5 years ago |
|
fill_op.cu.cc
|
add kernel for fill_op, test=develop (#19719)
|
5 years ago |
|
fill_op.h
|
Delete Ref & VectorRef and add GetDataSafely (#22997)
|
5 years ago |
|
fill_zeros_like_op.cc
|
rename inplace/no_need_buffer inferer, part2, test=develop (#24733)
|
5 years ago |
|
fill_zeros_like_op.cu.cc
|
…
|
|
|
fill_zeros_like_op.h
|
…
|
|
|
filter_by_instag_op.cc
|
Update paddle enforce message (#24498)
|
5 years ago |
|
filter_by_instag_op.h
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
flatten_op.cc
|
add new flatten op test=develop (#25393)
|
5 years ago |
|
flatten_op.cu.cc
|
add new flatten op test=develop (#25393)
|
5 years ago |
|
flatten_op.h
|
fix flatten api grad (#30426)
|
4 years ago |
|
flip_op.cc
|
add op_version for flip op [test=op_version] (#30019)
|
4 years ago |
|
flip_op.cu
|
modify flip test=develop (#25312)
|
5 years ago |
|
flip_op.h
|
modify flip test=develop (#25312)
|
5 years ago |
|
fsp_op.cc
|
Fix fsp_op error message,test=develop (#24405)
|
5 years ago |
|
fsp_op.cu
|
…
|
|
|
fsp_op.h
|
Remove disable flag in test_fsp_op.py (#22171)
|
5 years ago |
|
gather.cu.h
|
fix error message of gather nd (#29521)
|
4 years ago |
|
gather.h
|
add paddle.gather for API2.0 (#26455)
|
5 years ago |
|
gather_nd_op.cc
|
gather_nd Op for API 2.0 refine (#26540)
|
5 years ago |
|
gather_nd_op.cu
|
fix error message, test=develop (#24425)
|
5 years ago |
|
gather_nd_op.h
|
fix error message, test=develop (#24425)
|
5 years ago |
|
gather_op.cc
|
refine gather OP performance for dynamic mode (#28587)
|
4 years ago |
|
gather_op.cu
|
add paddle.gather for API2.0 (#26455)
|
5 years ago |
|
gather_op.h
|
add paddle.gather for API2.0 (#26455)
|
5 years ago |
|
gather_op_xpu.cc
|
add gather_op xpu, test=kunlun (#27822)
|
4 years ago |
|
gather_test.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
gather_tree_op.cc
|
Optimize the error message of OP. (#27478)
|
4 years ago |
|
gather_tree_op.cu
|
Fix index overflow bug of the CUDA kernel loop increment (#25435)
|
5 years ago |
|
gather_tree_op.h
|
Add seq2seq api related code (#19820)
|
5 years ago |
|
gaussian_random_batch_size_like_op.cc
|
rename inplace/no_need_buffer inferer, part2, test=develop (#24733)
|
5 years ago |
|
gaussian_random_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
gaussian_random_op.cu
|
add empty op (c++, python, unit test) (#26659)
|
4 years ago |
|
gaussian_random_op_xpu.cc
|
Add gaussian_random XPU kernels (#27853)
|
4 years ago |
|
gelu_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
gelu_op.cu
|
add approximation for gelu, test=develop (#22961)
|
5 years ago |
|
gelu_op.h
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
get_tensor_from_selected_rows_op.cc
|
[ROCM] update fluid operators for rocm (part7), test=develop (#31307)
|
4 years ago |
|
grid_sampler_cudnn_op.cu.cc
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
grid_sampler_op.cc
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
grid_sampler_op.cu
|
Fix round in grid sample op (#27657)
|
4 years ago |
|
grid_sampler_op.h
|
Make grid support stopping graients. (#27630)
|
4 years ago |
|
group_norm_op.cc
|
rename inplace/no_need_buffer inferer, part2, test=develop (#24733)
|
5 years ago |
|
group_norm_op.cu
|
[ROCM] fix dropout and remove hipcub, test=develop (#31455)
|
4 years ago |
|
group_norm_op.h
|
fixed group_norm's bug and modified unittest (#20506)
|
5 years ago |
|
gru_op.cc
|
rename inplace/no_need_buffer inferer, part2, test=develop (#24733)
|
5 years ago |
|
gru_op.cu.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
gru_op.h
|
…
|
|
|
gru_unit_op.cc
|
Move input_size check into RunTime phrase of gru_unit_op and refine error message (#24776)
|
5 years ago |
|
gru_unit_op.cu
|
…
|
|
|
gru_unit_op.h
|
optimize unity build (#31119)
|
4 years ago |
|
hash_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
hash_op.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
hierarchical_sigmoid_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
hierarchical_sigmoid_op.h
|
[Feature] one ps (3/4) (#29604)
|
4 years ago |
|
hinge_loss_op.cc
|
Error message enhancement of 6 op (#23759)
|
5 years ago |
|
hinge_loss_op.cu
|
…
|
|
|
hinge_loss_op.h
|
…
|
|
|
histogram_op.cc
|
Add histc op (#24562)
|
5 years ago |
|
histogram_op.cu
|
update histogram op for performance optimization, test=develop (#24912)
|
4 years ago |
|
histogram_op.h
|
Add histc op (#24562)
|
5 years ago |
|
huber_loss_op.cc
|
Error message enhancement of 6 op (#23759)
|
5 years ago |
|
huber_loss_op.cu
|
support fp64 in huber_loss cuda kernel (#26583)
|
5 years ago |
|
huber_loss_op.h
|
Add dygraph execution context (#20157)
|
5 years ago |
|
im2sequence_op.cc
|
Enhance checking in some operator. (#24473)
|
5 years ago |
|
im2sequence_op.cu
|
…
|
|
|
im2sequence_op.h
|
…
|
|
|
imag_op.cc
|
[Complex] Add real & imag op and api for complex tensor (#29672)
|
4 years ago |
|
imag_op.cu
|
[Complex] Add real & imag op and api for complex tensor (#29672)
|
4 years ago |
|
imag_op.h
|
[Complex] Add real & imag op and api for complex tensor (#29672)
|
4 years ago |
|
increment_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
increment_op.cu
|
…
|
|
|
increment_op.h
|
…
|
|
|
index_sample_op.cc
|
Update index sample (#24109)
|
5 years ago |
|
index_sample_op.cu
|
fix the bug in backward OP of index_sample. (#31026)
|
4 years ago |
|
index_sample_op.h
|
fix header file paths of gflags, commit 3, test=develop (#30273)
|
4 years ago |
|
index_select_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
index_select_op.cu
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
index_select_op.h
|
test=develop, bug fix for index_select and roll op (#25251)
|
5 years ago |
|
inplace_abn_op.cc
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
inplace_abn_op.cu
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
inplace_abn_op.h
|
Add inplace abn op (#22806)
|
5 years ago |
|
instance_norm_op.cc
|
register ModifyAttr for instance_norm, test=op_version (#30065)
|
4 years ago |
|
instance_norm_op.cu
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
instance_norm_op.h
|
improve efficiency of runtime InferVarType (#22778)
|
5 years ago |
|
interpolate_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
interpolate_op.cu
|
refine gpu kernel config for Paddle (#28085)
|
4 years ago |
|
interpolate_op.h
|
add linear interpolate operator (#23357)
|
5 years ago |
|
interpolate_op_xpu.cc
|
update activation op on kunlun (#29577)
|
4 years ago |
|
interpolate_v2_op.cc
|
Polish some error message in opeators (#27876)
|
4 years ago |
|
interpolate_v2_op.cu
|
refine gpu kernel config for Paddle (#28085)
|
4 years ago |
|
interpolate_v2_op.h
|
fix typo for interp_v2,test=develop (#26843)
|
4 years ago |
|
interpolate_v2_op_xpu.cc
|
add nearest_interp_v2 on kunlun (#29725)
|
4 years ago |
|
inverse_op.cc
|
Add the implementation of inverse (#23310)
|
5 years ago |
|
inverse_op.cu.cc
|
Add the implementation of inverse (#23310)
|
5 years ago |
|
inverse_op.h
|
Add the implementation of inverse (#23310)
|
5 years ago |
|
is_empty_op.cc
|
fix check and error message for flatten hash is_empty op (#24434)
|
5 years ago |
|
is_empty_op.cu.cc
|
…
|
|
|
is_empty_op.h
|
…
|
|
|
isfinite_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
isfinite_op.cu
|
…
|
|
|
isfinite_op.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
isfinite_v2_op.cc
|
fix isfinite_v2_op OpProtoAndCheckerMaker AddComment bug (#29626)
|
4 years ago |
|
isfinite_v2_op.cu
|
Add isfinite v2 op (#26344)
|
5 years ago |
|
isfinite_v2_op.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
kldiv_loss_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
kldiv_loss_op.cu
|
…
|
|
|
kldiv_loss_op.h
|
optimize unity build (#31119)
|
4 years ago |
|
kron_op.cc
|
type promotion for grad (#30177)
|
4 years ago |
|
kron_op.cu
|
Make transpose, trace, kron, reshape, sum op support complex type (#29321)
|
4 years ago |
|
kron_op.h
|
[ROCM] fix dropout and remove hipcub, test=develop (#31455)
|
4 years ago |
|
l1_norm_op.cc
|
API/OP error message enhancement (#23684)
|
5 years ago |
|
l1_norm_op.cu
|
…
|
|
|
l1_norm_op.h
|
API/OP error message enhancement (#23684)
|
5 years ago |
|
label_smooth_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
label_smooth_op.cu
|
Fix default label dim of label_smooth_op. test=develop (#21862)
|
5 years ago |
|
label_smooth_op.h
|
Fix default label dim of label_smooth_op. test=develop (#21862)
|
5 years ago |
|
layer_norm_op.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
layer_norm_op.cu
|
[ROCM] fix layer_norm, norm, p_norm, test_sequence_softmax_op, test_math_op_patch_var_base (#31709)
|
4 years ago |
|
layer_norm_op.h
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
layer_norm_op_xpu.cc
|
support transformer v2.0 (#30381)
|
4 years ago |
|
layout_utils.h
|
support channel last in BatchNorm*d
|
4 years ago |
|
linear_chain_crf_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
linear_chain_crf_op.h
|
optimize unity build (#31119)
|
4 years ago |
|
linspace_op.cc
|
Register op version for linspace,test=op_version (#30025)
|
4 years ago |
|
linspace_op.cu
|
refine the precious of linspace Op using half way (#27452)
|
4 years ago |
|
linspace_op.h
|
refine the precious of linspace Op using half way (#27452)
|
4 years ago |
|
load_combine_op.cc
|
…
|
|
|
load_combine_op.cu
|
…
|
|
|
load_combine_op.h
|
fix eigen in push sparse; fix hadoop command (#26872)
|
5 years ago |
|
load_op.cc
|
memory leak for cpu (#21174)
|
5 years ago |
|
load_op.cu
|
…
|
|
|
load_op.h
|
Op (Save/Load) error message enhancement, test=develop (#23650)
|
5 years ago |
|
load_op_xpu.cc
|
add load_op_xpu for Baidu Kunlun (#27817)
|
4 years ago |
|
lod_array_length_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
lod_rank_table_op.cc
|
enhance array_to_lod_tensor_op lod_tensor_to_array_op errors informaiton (#27386)
|
4 years ago |
|
lod_reset_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
lod_reset_op.cu
|
…
|
|
|
lod_reset_op.h
|
Api (lod_append) error message enhancement (#23541)
|
5 years ago |
|
lod_tensor_to_array_op.cc
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
log_loss_op.cc
|
API/OP (affine_channel, group_norm, layer_norm, random_crop, unpool, … (#24118)
|
5 years ago |
|
log_loss_op.cu
|
…
|
|
|
log_loss_op.h
|
…
|
|
|
log_loss_op_xpu.cc
|
Polish kunlun error (#27974)
|
4 years ago |
|
log_softmax_op.cc
|
log_softmax and LogSoftmax: impl kernel and refind docs (#26088)
|
5 years ago |
|
log_softmax_op.cu
|
log_softmax and LogSoftmax: impl kernel and refind docs (#26088)
|
5 years ago |
|
log_softmax_op.h
|
fix softmax cross entropy integer overflow (#30590)
|
4 years ago |
|
lookup_table_dequant_op.cc
|
add lookup_table_dequant_op (#22900)
|
5 years ago |
|
lookup_table_dequant_op.h
|
[Feature] one ps (3/4) (#29604)
|
4 years ago |
|
lookup_table_op.cc
|
[oneDNN] lookup_table op with support for BF16 data type. (#31558)
|
4 years ago |
|
lookup_table_op.cu
|
Bugfix rocm (#31490)
|
4 years ago |
|
lookup_table_op.h
|
[oneDNN] lookup_table op with support for BF16 data type. (#31558)
|
4 years ago |
|
lookup_table_v2_op.cc
|
enhance error messages of lookup_tale, merge_ids, data_norm (#27619)
|
4 years ago |
|
lookup_table_v2_op.cu
|
fix gpu outofrange (#29238)
|
4 years ago |
|
lookup_table_v2_op.h
|
[Feature] one ps (3/4) (#29604)
|
4 years ago |
|
lookup_table_v2_op_xpu.cc
|
add squeeze_op/unsqueeze_op on kunlun;fix conv op and parallel executor;optimize lookup_table op (#31056)
|
4 years ago |
|
lrn_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
lrn_op.cu
|
lrn supports channel_last input, test=develop (#20954)
|
5 years ago |
|
lrn_op.h
|
Error message enhancement of 6 op (#23759)
|
5 years ago |
|
lstm_op.cc
|
API(dynamic_lstm, dynamic_lstmp) error message enhancement (#24450)
|
5 years ago |
|
lstm_op.cu.cc
|
…
|
|
|
lstm_op.h
|
API(dynamic_gru, chunk_eval, BeamSearchDecoder) error message enhancement (#24513)
|
5 years ago |
|
lstm_unit_op.cc
|
API(lstm_unit, lstmp, sequence_mask, sequence_enumerate, sequence_conv) error message enhancement (#27572)
|
4 years ago |
|
lstm_unit_op.cu
|
API(lstm_unit, lstmp, sequence_mask, sequence_enumerate, sequence_conv) error message enhancement (#27572)
|
4 years ago |
|
lstm_unit_op.h
|
API(lstm_unit, lstmp, sequence_mask, sequence_enumerate, sequence_conv) error message enhancement (#27572)
|
4 years ago |
|
lstmp_op.cc
|
API(dynamic_lstm, dynamic_lstmp) error message enhancement (#24450)
|
5 years ago |
|
lstmp_op.cu
|
…
|
|
|
lstmp_op.h
|
Modify relu native implementation 2 (#30996)
|
4 years ago |
|
margin_rank_loss_op.cc
|
API/OP (margin_rank_loss, nce, row_conv, positive_negative_pair) erro… (#24246)
|
5 years ago |
|
margin_rank_loss_op.cu
|
…
|
|
|
margin_rank_loss_op.h
|
…
|
|
|
masked_select_op.cc
|
【API2.0】add masked_select Op for API2.0 (#26374)
|
5 years ago |
|
masked_select_op.cu
|
【API2.0】add masked_select Op for API2.0 (#26374)
|
5 years ago |
|
masked_select_op.h
|
【API2.0】add masked_select Op for API2.0 (#26374)
|
5 years ago |
|
match_matrix_tensor_op.cc
|
Refine error message of MatchMatrix and PyramidHash (#27484)
|
4 years ago |
|
match_matrix_tensor_op.h
|
Add match_matrix_tensor op (#18525)
|
6 years ago |
|
math.h
|
codegen for fused elementwise operation (#19520)
|
6 years ago |
|
matmul_op.cc
|
Polish two error messages (#31852)
|
4 years ago |
|
matmul_op_xpu.cc
|
opt matmul and matmul_v2 on kunlun, *test=kunlun (#31326)
|
4 years ago |
|
matmul_v2_op.cc
|
type promotion for grad (#30177)
|
4 years ago |
|
matmul_v2_op.cu
|
add complex64 and complex128 type; add +-*/@ and slice opreator for c… (#29199)
|
4 years ago |
|
matmul_v2_op.h
|
[ROCM] fix dropout and remove hipcub, test=develop (#31455)
|
4 years ago |
|
matmul_v2_op_xpu.cc
|
opt matmul and matmul_v2 on kunlun, *test=kunlun (#31326)
|
4 years ago |
|
max_sequence_len_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
maxout_op.cc
|
refine APIs: brelu, hardsigmoid, hardswish, maxout (#27658)
|
4 years ago |
|
maxout_op.cu.cc
|
…
|
|
|
maxout_op.h
|
refine APIs: brelu, hardsigmoid, hardswish, maxout (#27658)
|
4 years ago |
|
mean_iou_op.cc
|
OP Normal, Uniform, Xavier Initializer, smooth_l1, mean_iou error message enhancement (#23751)
|
5 years ago |
|
mean_iou_op.cu
|
Fix index overflow bug of the CUDA kernel loop increment (#25435)
|
5 years ago |
|
mean_iou_op.h
|
…
|
|
|
mean_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
mean_op.cu
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
mean_op.h
|
OP error message enhancement of l2_normalize, matmul, mean, etc
|
5 years ago |
|
mean_op_xpu.cc
|
error message optimization in mean_xpu,softmax_with_cross_entropy_op_xpu,test=kunlun (#27967)
|
4 years ago |
|
memcpy_op.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
memcpy_op.h
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
merge_lod_tensor_op.cc
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
merge_selected_rows_op.cc
|
Improving error reporting messages for ops (#24438)
|
5 years ago |
|
merge_selected_rows_op.cu.cc
|
…
|
|
|
merge_selected_rows_op.h
|
…
|
|
|
meshgrid_op.cc
|
Add meshgrid op (#23736)
|
5 years ago |
|
meshgrid_op.cu
|
Add meshgrid op (#23736)
|
5 years ago |
|
meshgrid_op.h
|
optimize unity build (#30195)
|
4 years ago |
|
minus_op.cc
|
OP(minus) error message enhancement. test=develop (#23621)
|
5 years ago |
|
minus_op.cu
|
…
|
|
|
minus_op.h
|
…
|
|
|
miopen_lstm_cache.h
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
miopen_rnn_cache.h
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
mish_op.cc
|
add mish op. (#24565)
|
5 years ago |
|
mish_op.cu
|
refine gpu kernel config for Paddle (#28085)
|
4 years ago |
|
mish_op.h
|
add mish op. (#24565)
|
5 years ago |
|
modified_huber_loss_op.cc
|
Error message enhancement of 6 op (#23759)
|
5 years ago |
|
modified_huber_loss_op.cu
|
…
|
|
|
modified_huber_loss_op.h
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
mul_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
mul_op.cu.cc
|
…
|
|
|
mul_op.h
|
Add mkldnn int8 mul-op kernel (#17834)
|
6 years ago |
|
mul_op_xpu.cc
|
support elementwise add, activation, matmul on Baidu Kunlun (#27143)
|
4 years ago |
|
multinomial_op.cc
|
Fix error message of multinomial op (#27946)
|
4 years ago |
|
multinomial_op.cu
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
multinomial_op.h
|
Fix error message of multinomial op (#27946)
|
4 years ago |
|
multiplex_op.cc
|
Upgrade Error Message for AucOP & MultiplexOP (#24458)
|
5 years ago |
|
multiplex_op.cu
|
Upgrade Error Message for AucOP & MultiplexOP (#24458)
|
5 years ago |
|
multiplex_op.h
|
Upgrade Error Message for AucOP & MultiplexOP (#24458)
|
5 years ago |
|
mv_op.cc
|
update mv op according PR#27024 (#27474)
|
4 years ago |
|
mv_op.cu
|
refine gpu kernel config for Paddle (#28085)
|
4 years ago |
|
mv_op.h
|
update mv op according PR#27024 (#27474)
|
4 years ago |
|
nce_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
nce_op.h
|
[Feature] one ps (3/4) (#29604)
|
4 years ago |
|
nll_loss_op.cc
|
enhance error message of nll_loss op test=develop (#30125)
|
4 years ago |
|
nll_loss_op.cu
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
nll_loss_op.h
|
Polish two error messages (#31852)
|
4 years ago |
|
norm_op.cc
|
API/OP error message enhancement (#23684)
|
5 years ago |
|
norm_op.cu
|
[ROCM] fix layer_norm, norm, p_norm, test_sequence_softmax_op, test_math_op_patch_var_base (#31709)
|
4 years ago |
|
norm_op.h
|
…
|
|
|
norm_utils.cu.h
|
[ROCM] update fluid operators for rocm (part8), test=develop (#31309)
|
4 years ago |
|
norm_utils.h
|
add instance norm (#19500)
|
5 years ago |
|
one_hot_op.cc
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
one_hot_op.cu
|
supports collective communicated training (#18175)
|
6 years ago |
|
one_hot_op.h
|
Error message enhancement of 6 op (#23759)
|
5 years ago |
|
one_hot_op_xpu.cc
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
one_hot_v2_op.cc
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
one_hot_v2_op.cu
|
Remove constraint that last dimension is forced to be 1 by adding one_hot_v2 (#19716)
|
5 years ago |
|
one_hot_v2_op.h
|
Update paddle enforce message (#24498)
|
5 years ago |
|
one_hot_v2_op_xpu.cc
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
op_debug_string_test.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
p_norm_op.cc
|
Add p_norm op version info (#30042)
|
4 years ago |
|
p_norm_op.cu
|
[ROCM] fix layer_norm, norm, p_norm, test_sequence_softmax_op, test_math_op_patch_var_base (#31709)
|
4 years ago |
|
p_norm_op.h
|
Norm op support 2-axis (#26492)
|
5 years ago |
|
pad2d_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
pad2d_op.cu
|
Fix index overflow bug of the CUDA kernel loop increment (#25435)
|
5 years ago |
|
pad3d_op.cc
|
add pad and concat double grad (#29549)
|
4 years ago |
|
pad3d_op.cu
|
add pad func (#26106)
|
5 years ago |
|
pad_constant_like_op.cc
|
OP(pad, pad2d, pad_constant_like) error message enhancement (#23882)
|
5 years ago |
|
pad_constant_like_op.cu
|
add register op_data_type of pad/expand_as et.al (#21718)
|
5 years ago |
|
pad_constant_like_op.h
|
Update the precision of pad, pad2d, pad_constant_like's unit tests from fp32 to fp64 (#22394)
|
5 years ago |
|
pad_op.cc
|
add pad and concat double grad (#29549)
|
4 years ago |
|
pad_op.cu
|
add register op_data_type of pad/expand_as et.al (#21718)
|
5 years ago |
|
pad_op.h
|
Add fp16 support for pad and split (#19881)
|
5 years ago |
|
partial_concat_op.cc
|
Imperative tracer refactoring (#22457)
|
5 years ago |
|
partial_concat_op.cu
|
Add macro BOOST_GET to enrich the error information of boost :: get (#24175)
|
5 years ago |
|
partial_concat_op.h
|
add partial_concat op in contrib (#22528)
|
5 years ago |
|
partial_sum_op.cc
|
Imperative tracer refactoring (#22457)
|
5 years ago |
|
partial_sum_op.cu
|
Add macro BOOST_GET to enrich the error information of boost :: get (#24175)
|
5 years ago |
|
partial_sum_op.h
|
add partial_sum op in contrib (#22292)
|
5 years ago |
|
pixel_shuffle_op.cc
|
Add version checking, test=op_version (#30129)
|
4 years ago |
|
pixel_shuffle_op.cu
|
…
|
|
|
pixel_shuffle_op.h
|
[Api2.0] add pixel shuffle (#26071)
|
5 years ago |
|
pool_cudnn_op.cu.cc
|
[ROCM] update fluid operators for rocm (part4), test=develop (#31225)
|
4 years ago |
|
pool_op.cc
|
[ROCM] update fluid operators for rocm (part4), test=develop (#31225)
|
4 years ago |
|
pool_op.cu
|
Optimized the adaptive_avg_pool2d op when output_size == 1 (#31197)
|
4 years ago |
|
pool_op.h
|
[ROCM] fix dropout and remove hipcub, test=develop (#31455)
|
4 years ago |
|
pool_op_xpu.cc
|
optimize batch_norm & pool op for kunlun (#30490)
|
4 years ago |
|
pool_with_index_op.cc
|
Error message opt, test=develop (#27467)
|
4 years ago |
|
pool_with_index_op.cu.cc
|
…
|
|
|
pool_with_index_op.h
|
Error message opt, test=develop (#27467)
|
4 years ago |
|
positive_negative_pair_op.cc
|
fix error mesage for negative_positive_pair_op and nce_op (#27779)
|
4 years ago |
|
positive_negative_pair_op.h
|
…
|
|
|
prelu_op.cc
|
fix bug of prelu when rank not equal 4, test=develop (#25067)
|
5 years ago |
|
prelu_op.cu
|
[ROCM] fix dropout and remove hipcub, test=develop (#31455)
|
4 years ago |
|
prelu_op.h
|
fix the computation for dx (grad for x) for prelu operation. (#20949)
|
5 years ago |
|
print_op.cc
|
Register op version for print, test=op_version (#29945)
|
4 years ago |
|
prroi_pool_op.cc
|
API/OP (center_loss, fluid.one_hot, prroi_pool, roi_pool, ctc_greed_decoder) error message enhancement (#23794)
|
5 years ago |
|
prroi_pool_op.cu
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
prroi_pool_op.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
psroi_pool_op.cc
|
Update paddle enforce message (#24498)
|
5 years ago |
|
psroi_pool_op.cu
|
Error message opt, test=develop (#27467)
|
4 years ago |
|
psroi_pool_op.h
|
Update paddle enforce message (#24498)
|
5 years ago |
|
pull_box_extended_sparse_op.cc
|
fix conflict, test=develop (#24238)
|
5 years ago |
|
pull_box_extended_sparse_op.cu
|
fix conflict, test=develop (#24238)
|
5 years ago |
|
pull_box_extended_sparse_op.h
|
fix conflict, test=develop (#24238)
|
5 years ago |
|
pull_box_sparse_op.cc
|
heter box (#29734)
|
4 years ago |
|
pull_box_sparse_op.cu
|
Paddlebox Framework (#18982)
|
6 years ago |
|
pull_box_sparse_op.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
pull_sparse_op.cc
|
add fleet pslib pull and push sparse op and push dense op (#23139)
|
5 years ago |
|
pull_sparse_op.h
|
add fleet pslib pull and push sparse op and push dense op (#23139)
|
5 years ago |
|
pull_sparse_v2_op.cc
|
add fleet pslib pull and push sparse op and push dense op (#23139)
|
5 years ago |
|
pull_sparse_v2_op.h
|
add fleet pslib pull and push sparse op and push dense op (#23139)
|
5 years ago |
|
push_dense_op.cc
|
rename inplace/no_need_buffer inferer, part 1, test=develop (#24711)
|
5 years ago |
|
push_dense_op.h
|
add fleet pslib pull and push sparse op and push dense op (#23139)
|
5 years ago |
|
py_func_op.cc
|
enhance error info for py_func (#30138)
|
4 years ago |
|
py_func_op.h
|
…
|
|
|
pyramid_hash_op.cc
|
Refine error message of MatchMatrix and PyramidHash (#27484)
|
4 years ago |
|
quantize_op.cc
|
operator checkpoints for new attributes. (#29832)
|
4 years ago |
|
quantize_op.h
|
…
|
|
|
queue_generator_op.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
randint_op.cc
|
Refine paddle.manual_seed (#26496)
|
5 years ago |
|
randint_op.cu
|
add cuda generator (#26786)
|
5 years ago |
|
random_crop_op.cc
|
API/OP (affine_channel, group_norm, layer_norm, random_crop, unpool, … (#24118)
|
5 years ago |
|
random_crop_op.cu
|
…
|
|
|
random_crop_op.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
randperm_op.cc
|
Fix the formate of raising error in randperm op (#30108)
|
4 years ago |
|
randperm_op.cu
|
randperm API: remove out, devive, stop_gradient; add name (#25410)
|
5 years ago |
|
randperm_op.h
|
randperm run error in multi-gpus (#27942)
|
4 years ago |
|
range_op.cc
|
avoid data transfer, test=develop (#25810)
|
5 years ago |
|
range_op.cu
|
optimize range op by place parameters on cpu rather than gpu, test=develop (#30811)
|
4 years ago |
|
range_op.h
|
fix error log, test=develop (#24419)
|
5 years ago |
|
range_op_xpu.cc
|
dyngraph (#30892)
|
4 years ago |
|
rank_attention.cu.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
rank_attention_op.cc
|
fix error message (#30135)
|
4 years ago |
|
rank_attention_op.cu
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
rank_attention_op.h
|
fix conflict, test=develop (#23298)
|
5 years ago |
|
rank_loss_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
rank_loss_op.cu
|
…
|
|
|
rank_loss_op.h
|
optimize unity build (#30195)
|
4 years ago |
|
real_op.cc
|
[Complex] Add real & imag op and api for complex tensor (#29672)
|
4 years ago |
|
real_op.cu
|
[Complex] Add real & imag op and api for complex tensor (#29672)
|
4 years ago |
|
real_op.h
|
[Complex] Add real & imag op and api for complex tensor (#29672)
|
4 years ago |
|
recurrent_op.cc
|
fix runtime crash when rnn model inference, test=develop (#31833)
|
4 years ago |
|
recurrent_op.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
reorder_lod_tensor_by_rank_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
requantize_op.cc
|
operator checkpoints for new attributes. (#29832)
|
4 years ago |
|
requantize_op.h
|
…
|
|
|
reshape_op.cc
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
reverse_op.cc
|
Support LoDTensorArray in reverse_op (#24797)
|
5 years ago |
|
reverse_op.cu
|
…
|
|
|
reverse_op.h
|
Support LoDTensorArray in reverse_op (#24797)
|
5 years ago |
|
rnn_memory_helper_op.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
rnn_op.cc
|
Add LSTM, Simple RNN and GRU CPU kernel (#28577)
|
4 years ago |
|
rnn_op.cu.cc
|
[ROCM] fix test_rnn_op (#31735)
|
4 years ago |
|
rnn_op.h
|
Modify relu native implementation 2 (#30996)
|
4 years ago |
|
roi_align_op.cc
|
add offset parameter in roi_align,generate_proposals.etc ops (#30864)
|
4 years ago |
|
roi_align_op.cu
|
fix roi_align, test=develop (#31479)
|
4 years ago |
|
roi_align_op.h
|
fix roi_align, test=develop (#31479)
|
4 years ago |
|
roi_align_op_xpu.cc
|
support roi_align & affine_channel for kunlun (#29561)
|
4 years ago |
|
roi_pool_op.cc
|
add REGISTER_OP_VERSION for generate_proposals, roi_align, roi_pool test=op_version (#30034)
|
4 years ago |
|
roi_pool_op.cu
|
Error message opt, test=develop (#27467)
|
4 years ago |
|
roi_pool_op.h
|
Enhance ops to support LoD as input for dygraph detection models. (#25316)
|
4 years ago |
|
roll_op.cc
|
test=develop, add op_register_version for roll_op (#30023)
|
4 years ago |
|
roll_op.cu
|
Roll cuda kernel (#29655)
|
4 years ago |
|
roll_op.h
|
modify roll test=develop (#25321)
|
5 years ago |
|
row_conv_op.cc
|
API/OP (margin_rank_loss, nce, row_conv, positive_negative_pair) erro… (#24246)
|
5 years ago |
|
row_conv_op.cu
|
fix row_conv_op to force it support lodtensor and tensor input simultaneously, test=develop (#19412)
|
6 years ago |
|
row_conv_op.h
|
…
|
|
|
run_program_op.cc
|
fix loaded no params layer run error (#27241)
|
4 years ago |
|
run_program_op.cu.cc
|
Implement StaticModelRunner to support dygraph fine-tune static graph pre-training model (#23171)
|
5 years ago |
|
run_program_op.h
|
[Dy2Stat] Add cache for Executor and Context in run_program_op (#28421)
|
4 years ago |
|
sample_logits_op.cc
|
Remove extraneous comma in error messages (#24478)
|
5 years ago |
|
sample_logits_op.cu
|
fix lod_reset bug, test=develop (#21392)
|
5 years ago |
|
sample_logits_op.h
|
Remove extraneous comma in error messages (#24478)
|
5 years ago |
|
sampling_id_op.cc
|
SamplingID Op fix error print (#24521)
|
5 years ago |
|
sampling_id_op.cu
|
…
|
|
|
sampling_id_op.h
|
Refine paddle.manual_seed (#26496)
|
5 years ago |
|
save_combine_op.cc
|
improve efficiency of runtime InferVarType (#22778)
|
5 years ago |
|
save_combine_op.cu
|
add register op_data_type of pad/expand_as et.al (#21718)
|
5 years ago |
|
save_combine_op.h
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
save_load_combine_op_test.cc
|
…
|
|
|
save_load_op_test.cc
|
…
|
|
|
save_op.cc
|
Incorporate cudnn_lstm into LSTM api (#27217)
|
4 years ago |
|
save_op.cu
|
Incorporate cudnn_lstm into LSTM api (#27217)
|
4 years ago |
|
save_op.h
|
delete include framework.pb.h (#31859)
|
4 years ago |
|
scale_op.cc
|
[oneDNN] Initial bf16 amp integration (#31093)
|
4 years ago |
|
scale_op.cu
|
refine math_op_patch, test=develop (#19727)
|
6 years ago |
|
scale_op.h
|
add the error message check for the some operator
|
4 years ago |
|
scale_op_xpu.cc
|
support transformer v2.0 (#30381)
|
4 years ago |
|
scatter.cu.h
|
Fix scatter grad bug (#30604)
|
4 years ago |
|
scatter.h
|
Fix scatter grad bug (#30604)
|
4 years ago |
|
scatter_nd_add_op.cc
|
rename inplace/no_need_buffer inferer, part4, test=develop (#24781)
|
5 years ago |
|
scatter_nd_add_op.cu
|
Fix scatter grad bug (#30604)
|
4 years ago |
|
scatter_nd_add_op.h
|
Fix scatter grad bug (#30604)
|
4 years ago |
|
scatter_op.cc
|
Fix scatter grad bug (#30604)
|
4 years ago |
|
scatter_op.cu
|
Fix scatter grad bug (#30604)
|
4 years ago |
|
scatter_op.h
|
Fix scatter grad bug (#30604)
|
4 years ago |
|
scatter_test.cc
|
use iwyu clean include (#27267)
|
4 years ago |
|
search_compute.h
|
Support mips arch (#29903)
|
4 years ago |
|
seed_op.cc
|
add cuda kernel for seed, test=develop (#23749)
|
5 years ago |
|
seed_op.cu
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
seed_op.h
|
Dropout with seed (#21590)
|
5 years ago |
|
segment_pool_op.cc
|
Add the cpu version of segment sum mean max min op
|
4 years ago |
|
segment_pool_op.cu
|
refine gpu kernel config for Paddle (#28085)
|
4 years ago |
|
segment_pool_op.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
select_input_op.cc
|
Move input_size check into RunTime phrase of gru_unit_op and refine error message (#24776)
|
5 years ago |
|
select_op_helper.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
select_output_op.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
selu_op.cc
|
[OpDevOptimize] Add common infershape functions (#26096)
|
5 years ago |
|
selu_op.cu
|
…
|
|
|
selu_op.h
|
…
|
|
|
set_value_op.cc
|
Fix bug of set_value op：Decerease axes to do right broadcast (#31875)
|
4 years ago |
|
set_value_op.cu
|
[setitem] Support Tensor setitem in static mode (#29708)
|
4 years ago |
|
set_value_op.h
|
Fix bug of set_value op：Decerease axes to do right broadcast (#31875)
|
4 years ago |
|
shape_op.cc
|
shape op support int8 and uint8 tensor (#30201)
|
4 years ago |
|
shape_op.cu
|
shape op support int8 and uint8 tensor (#30201)
|
4 years ago |
|
shape_op.h
|
[Dy2stat] Support len syntax (#24638)
|
5 years ago |
|
shape_op_xpu.cc
|
add XPU support for shape op and reshape op (#27804)
|
4 years ago |
|
shard_index_op.cc
|
Improving error reporting messages for ops (#24438)
|
5 years ago |
|
shard_index_op.cu
|
Improving error reporting messages for ops (#24438)
|
5 years ago |
|
shard_index_op.h
|
Improving error reporting messages for ops (#24438)
|
5 years ago |
|
shrink_rnn_memory_op.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
shuffle_batch_op.cc
|
Imperative tracer refactoring (#22457)
|
5 years ago |
|
shuffle_batch_op.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
shuffle_channel_op.cc
|
[2.0RC]refine error message in shuffle channel OP (#27505)
|
4 years ago |
|
shuffle_channel_op.cu
|
…
|
|
|
shuffle_channel_op.h
|
…
|
|
|
sigmoid_cross_entropy_with_logits_op.cc
|
Enhance error message of cross_entropy_op, sigmoid_cross_entropy_with_logits_op (#24485)
|
5 years ago |
|
sigmoid_cross_entropy_with_logits_op.cu
|
[ROCM] fix gather_op, sigmoid_cross_entropy_with_logits_op, test=develop (#31467)
|
4 years ago |
|
sigmoid_cross_entropy_with_logits_op.h
|
…
|
|
|
sign_op.cc
|
update error info of ops，add some test cases for raise message (#23750)
|
5 years ago |
|
sign_op.cu
|
…
|
|
|
sign_op.h
|
…
|
|
|
sign_op_xpu.cc
|
Polish some error message in opeators (#27876)
|
4 years ago |
|
similarity_focus_op.cc
|
OP(rank_loss, similarity_focus, squeeze) error message enhancement (#24448)
|
5 years ago |
|
similarity_focus_op.h
|
OP(rank_loss, similarity_focus, squeeze) error message enhancement (#24448)
|
5 years ago |
|
size_op.cc
|
fix gpu kernel for numel Op (#27085)
|
4 years ago |
|
size_op.cu
|
fix gpu kernel for numel Op (#27085)
|
4 years ago |
|
size_op.h
|
fix gpu kernel for numel Op (#27085)
|
4 years ago |
|
slice_op.cc
|
Add error message for slice op(#30851)
|
4 years ago |
|
slice_op.cu
|
add complex64 and complex128 type; add +-*/@ and slice opreator for c… (#29199)
|
4 years ago |
|
slice_op.h
|
Fix bug: In dynamic mode, if start or end is negetive, __getitem__ return wrong result(#30003)
|
4 years ago |
|
slice_op_xpu.cc
|
add kunlun kernel: slice, slice_grad, top_k, cast. *test=kunlun (#28542)
|
4 years ago |
|
smooth_l1_loss_op.cc
|
update enhance error message for Initializer, smooth_l1 (#23912)
|
5 years ago |
|
smooth_l1_loss_op.cu
|
…
|
|
|
smooth_l1_loss_op.h
|
…
|
|
|
softmax_cudnn_op.cu
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
softmax_op.cc
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
softmax_op.cu.cc
|
…
|
|
|
softmax_op.h
|
…
|
|
|
softmax_op_xpu.cc
|
fix softmax bug for multi_card in kunlun (#30600)
|
4 years ago |
|
softmax_with_cross_entropy_op.cc
|
add softmax_switch for softmax_with_cross_entropy_op, test=develop (#31428)
|
4 years ago |
|
softmax_with_cross_entropy_op.cu
|
[ROCM] fix softmax_with_cross_entropy_op, test=develop (#31629)
|
4 years ago |
|
softmax_with_cross_entropy_op.h
|
add softmax_switch for softmax_with_cross_entropy_op, test=develop (#31428)
|
4 years ago |
|
softmax_with_cross_entropy_op_xpu.cc
|
1. fix elementwise ops'bug 2. fix softmax_with_cross_entropy_op 3. add biliner_interp_op (#29448)
|
4 years ago |
|
space_to_depth_op.cc
|
op error info (#27856)
|
4 years ago |
|
space_to_depth_op.cu
|
add register op_data_type of pad/expand_as et.al (#21718)
|
5 years ago |
|
space_to_depth_op.h
|
…
|
|
|
spectral_norm_op.cc
|
Update OP_INOUT_CHECK (#23757)
|
5 years ago |
|
spectral_norm_op.cu
|
…
|
|
|
spectral_norm_op.h
|
fix PADDLE_THROW in spectral_norm_op.h. test=develop (#24414)
|
5 years ago |
|
split_lod_tensor_op.cc
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
split_op.cc
|
op error info (#27856)
|
4 years ago |
|
split_op.cu.cc
|
refine the split op for API 2.0 test=develop (#25320)
|
5 years ago |
|
split_op.h
|
op error info (#27856)
|
4 years ago |
|
split_selected_rows_op.cc
|
test=develop, error info improvement (#24496)
|
5 years ago |
|
split_selected_rows_op.cu
|
…
|
|
|
split_selected_rows_op.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
spp_op.cc
|
Update paddle enforce message (#24498)
|
5 years ago |
|
spp_op.cu.cc
|
…
|
|
|
spp_op.h
|
[ROCM] update fluid operators for rocm (part4), test=develop (#31225)
|
4 years ago |
|
squared_l2_distance_op.cc
|
rename inplace/no_need_buffer inferer, part4, test=develop (#24781)
|
5 years ago |
|
squared_l2_distance_op.cu
|
…
|
|
|
squared_l2_distance_op.h
|
optimize unity build (#30195)
|
4 years ago |
|
squared_l2_norm_op.cc
|
API/OP error message enhancement (#23684)
|
5 years ago |
|
squared_l2_norm_op.cu
|
…
|
|
|
squared_l2_norm_op.h
|
API/OP error message enhancement (#23684)
|
5 years ago |
|
squeeze_op.cc
|
add uint8 support for squeeze operator (#28734)
|
4 years ago |
|
squeeze_op.cu.cc
|
add uint8 support for squeeze operator (#28734)
|
4 years ago |
|
squeeze_op.h
|
add uint8 support for squeeze operator (#28734)
|
4 years ago |
|
squeeze_op_xpu.cc
|
add squeeze_op/unsqueeze_op on kunlun;fix conv op and parallel executor;optimize lookup_table op (#31056)
|
4 years ago |
|
stack_op.cc
|
Imperative tracer refactoring (#22457)
|
5 years ago |
|
stack_op.cu
|
refine gpu kernel config for Paddle (#28085)
|
4 years ago |
|
stack_op.h
|
Refine stack op to improve xlnet performance, test=develop (#22142)
|
5 years ago |
|
stack_op_xpu.cc
|
feat: support check_nan_inf for kunlun/xpu device (#29694)
|
4 years ago |
|
strided_memcpy.h
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
strided_memcpy_test.cc
|
[ROCM] update fluid operators for rocm (part9), test=develop (#31338)
|
4 years ago |
|
strided_slice_op.cc
|
add complex64 and complex128 type; add +-*/@ and slice opreator for c… (#29199)
|
4 years ago |
|
strided_slice_op.cu
|
add complex64 and complex128 type; add +-*/@ and slice opreator for c… (#29199)
|
4 years ago |
|
strided_slice_op.h
|
Support int32 int64 and fix bug (#24407)
|
5 years ago |
|
sum_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
sum_op.cu
|
update the error message check for the some ops
|
4 years ago |
|
sum_op.h
|
update the error message check for the some ops
|
4 years ago |
|
sum_op_xpu.cc
|
fix enforce msg of sum xpu op (#30113)
|
4 years ago |
|
sync_batch_norm_op.cc
|
Add macro BOOST_GET to enrich the error information of boost :: get (#24175)
|
5 years ago |
|
sync_batch_norm_op.cu
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
sync_batch_norm_op.cu.h
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
tdm_child_op.cc
|
Add Tdm child OP in contrib (#23241)
|
5 years ago |
|
tdm_child_op.h
|
fix header file paths of gflags, commit 3, test=develop (#30273)
|
4 years ago |
|
tdm_sampler_op.cc
|
Add Tdm sampler op in Contrib (#23290)
|
5 years ago |
|
tdm_sampler_op.h
|
fix header file paths of gflags, commit 3, test=develop (#30273)
|
4 years ago |
|
teacher_student_sigmoid_loss_op.cc
|
Refine error message, test=develop (#23823)
|
5 years ago |
|
teacher_student_sigmoid_loss_op.h
|
…
|
|
|
temporal_shift_op.cc
|
support NHWC for temporal_shift op (#31642)
|
4 years ago |
|
temporal_shift_op.cu
|
support NHWC for temporal_shift op (#31642)
|
4 years ago |
|
temporal_shift_op.h
|
support NHWC for temporal_shift op (#31642)
|
4 years ago |
|
tensor_array_to_tensor_op.cc
|
c++ API ( average_accumulates, tensor_array_to_tensor and average_accumulates) error message enhance. test=develop (#23631)
|
5 years ago |
|
tensor_formatter.cc
|
use iwyu clean include second time, test=develop (#30829)
|
4 years ago |
|
tensor_formatter.h
|
use iwyu clean include (#27267)
|
4 years ago |
|
test_common_infer_shape_functions.cc
|
[OpDevOptimize] Add common infershape functions (#26096)
|
5 years ago |
|
test_leaky_relu_grad_grad_functor.cc
|
fix leaky_relu op when alpha is zero, test=develop (#19833)
|
5 years ago |
|
test_leaky_relu_grad_grad_functor.cu
|
fix leaky_relu op when alpha is zero, test=develop (#19833)
|
5 years ago |
|
test_leaky_relu_grad_grad_functor.h
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
tile_op.cc
|
fix shape of tile_grad op (#29289)
|
4 years ago |
|
tile_op.cu
|
[API 2.0] add paddle.tile op (#26245)
|
5 years ago |
|
tile_op.h
|
fix shape of tile_grad op (#29289)
|
4 years ago |
|
top_k_function_cuda.h
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
top_k_op.cc
|
Polish some error message in opeators (#27876)
|
4 years ago |
|
top_k_op.cu
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
top_k_op.h
|
optimize unity build (#31119)
|
4 years ago |
|
top_k_op_xpu.cc
|
add kunlun kernel: slice, slice_grad, top_k, cast. *test=kunlun (#28542)
|
4 years ago |
|
top_k_v2_op.cc
|
update the code for the topk message optimize
|
4 years ago |
|
top_k_v2_op.cu
|
optimize topk op through limit SortTopK kernel entrance, test=develop (#30403)
|
4 years ago |
|
top_k_v2_op.h
|
optimize unity build (#31119)
|
4 years ago |
|
trace_op.cc
|
Optimize the error message of framework. (#30134)
|
4 years ago |
|
trace_op.cu
|
[ROCM] fix dropout and remove hipcub, test=develop (#31455)
|
4 years ago |
|
trace_op.h
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
transpose_op.cc
|
More precise mkldnn kernel rules in GetExpectedKernelType (#29840)
|
4 years ago |
|
transpose_op.cu
|
Make transpose, trace, kron, reshape, sum op support complex type (#29321)
|
4 years ago |
|
transpose_op.h
|
enhance reduce op which can reduce tensor with arbitrary rank
|
4 years ago |
|
transpose_op_xpu.cc
|
fix expand/uniform_random && concat/transpose to new api on xpu (#29280)
|
4 years ago |
|
tree_conv_op.cc
|
test=develop, error message of tree_conv OP enhancement (#23574)
|
5 years ago |
|
tree_conv_op.cu
|
…
|
|
|
tree_conv_op.h
|
…
|
|
|
tril_triu_op.cc
|
add fp16 support for tril_triu op (#30186)
|
4 years ago |
|
tril_triu_op.cu
|
add fp16 support for tril_triu op (#30186)
|
4 years ago |
|
tril_triu_op.h
|
add fp16 support for tril_triu op (#30186)
|
4 years ago |
|
truncated_gaussian_random_op.cc
|
Add truncated_gaussian_random XPU kernel (#27861)
|
4 years ago |
|
truncated_gaussian_random_op.cu
|
fix truncated_gaussian seed (#28777)
|
4 years ago |
|
truncated_gaussian_random_op.h
|
Add truncated_gaussian_random XPU kernel (#27861)
|
4 years ago |
|
truncated_gaussian_random_op_xpu.cc
|
Add truncated_gaussian_random XPU kernel (#27861)
|
4 years ago |
|
unbind_op.cc
|
add unbind op (#23359)
|
5 years ago |
|
unbind_op.cu.cc
|
add unbind op (#23359)
|
5 years ago |
|
unbind_op.h
|
add unbind op (#23359)
|
5 years ago |
|
unfold_op.cc
|
rename inplace/no_need_buffer inferer, part3, test=develop (#24734)
|
5 years ago |
|
unfold_op.cu
|
…
|
|
|
unfold_op.h
|
Polish PADDLE_ENFORCE of unfold_op (#24423)
|
5 years ago |
|
uniform_random_batch_size_like_op.cc
|
rename inplace/no_need_buffer inferer, part2, test=develop (#24733)
|
5 years ago |
|
uniform_random_op.cc
|
update the error message check for the some ops
|
4 years ago |
|
uniform_random_op.cu
|
update the error message check for the some ops
|
4 years ago |
|
uniform_random_op.h
|
update the error message check for the some ops
|
4 years ago |
|
uniform_random_op_xpu.cc
|
fix expand/uniform_random && concat/transpose to new api on xpu (#29280)
|
4 years ago |
|
unique_op.cc
|
fix a bug in op_version_registry, test=develop, test=op_version (#29994)
|
4 years ago |
|
unique_op.cu
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
unique_op.h
|
add dtype for unique (#26655)
|
5 years ago |
|
unique_with_counts_op.cc
|
Add some error meesage and dtyp, dtyep check for some ops (#23762)
|
5 years ago |
|
unique_with_counts_op.h
|
Add the op of unique_with_counts, expand count function of the op unique (#18720)
|
6 years ago |
|
unity_build_rule.cmake
|
optimize unity build (#31119)
|
4 years ago |
|
unpool_op.cc
|
API/OP (group_norm, layer_norm, random_crop, unpool) error message enhancement (#24413)
|
5 years ago |
|
unpool_op.cu.cc
|
…
|
|
|
unpool_op.h
|
…
|
|
|
unsqueeze_op.cc
|
add uint8 support for squeeze operator (#28734)
|
4 years ago |
|
unsqueeze_op.cu.cc
|
add uint8 support for squeeze operator (#28734)
|
4 years ago |
|
unsqueeze_op.h
|
Update paddle enforce message (#24498)
|
5 years ago |
|
unsqueeze_op_xpu.cc
|
add squeeze_op/unsqueeze_op on kunlun;fix conv op and parallel executor;optimize lookup_table op (#31056)
|
4 years ago |
|
unstack_op.cc
|
update error message for unstack op and lamb op; test=develop (#24439)
|
5 years ago |
|
unstack_op.cu
|
add kernel for unstack_op, test=develop (#19538)
|
5 years ago |
|
unstack_op.h
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
utils.h
|
xpu support for fill_constant Op (#27675)
|
4 years ago |
|
var_conv_2d_op.cc
|
Polish no onwer ops error message (#27448)
|
4 years ago |
|
var_conv_2d_op.h
|
Add var_conv_2d op (#18518)
|
6 years ago |
|
warpctc_op.cc
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
warpctc_op.cu.cc
|
add support to float64 input of warpctc op. (#27399)
|
4 years ago |
|
warpctc_op.h
|
[ROCM] update fluid platform for rocm (part5), test=develop (#31315)
|
4 years ago |
|
where_index_op.cc
|
add new api for Paddle2.0: nonzero, index_selct, roll, cross (#23176)
|
5 years ago |
|
where_index_op.cu
|
add new api for Paddle2.0: nonzero, index_selct, roll, cross (#23176)
|
5 years ago |
|
where_index_op.h
|
add new api for Paddle2.0: nonzero, index_selct, roll, cross (#23176)
|
5 years ago |
|
where_op.cc
|
rename inplace/no_need_buffer inferer, part3, test=develop (#24734)
|
5 years ago |
|
where_op.cu
|
refine gpu kernel config for Paddle (#28085)
|
4 years ago |
|
where_op.h
|
Implement a new C++ operator where and API tensor.where (#23220)
|
5 years ago |
niuliling123