- 11 10月, 2022 9 次提交
-
-
由 yeliang2258 提交于
* fix slice bugs * fix * update code * fix * update code
-
由 HongyuJia 提交于
-
由 HongyuJia 提交于
* refine conv_op mkldnn code * fix customized_type_value
-
由 HongyuJia 提交于
* refine sum_op mkldnn code * refine sum_op mkldnn code
-
由 HongyuJia 提交于
-
由 傅剑寒 提交于
* add fp16 data type for set_value * cancel flip modification * add fp16 dtype support for set_value
-
由 HongyuJia 提交于
* solve transpose2, follow #22402 * fix CI cmake * update REGISTER_OP_KERNEL of transpose2
-
由 HongyuJia 提交于
-
由 wuyefeilin 提交于
-
- 10 10月, 2022 8 次提交
-
-
由 YuanRisheng 提交于
* add yaml entry for rnn and rrnn_grad, move infershape function for rnn_grad to phi infer meta * WIP: move rnn kernrl to phi * Change the code generation to avoid converting from intializer list to tuple of heterogeneous types. This is only triggered when an api has intermediate outputs, and the result of the outputs are of heterogeneous types. * fix the bug that when none in a vector of tensors requires gradient, the conversion to InferShapeContext to InferMetaContext (a.k.a. BuildInferMetaContext) produces errorous results. * fix ci bugs * fix ci bugs * fix ci bugs * modify code according comment Co-authored-by: Nchenfeiyu <chenfeiyu@baidu.com>
-
由 Zhang Jun 提交于
-
由 carryyu 提交于
make fused_multi_transformer support dynamically set the cache_kvs' shape and support input prefix_caches. (#46777) * make fused_multi_transformer support dynamically set the cache_kvs' shape and support input prefix_caches.
-
由 HongyuJia 提交于
-
由 HongyuJia 提交于
-
由 HongyuJia 提交于
* delete_quantize_headerfile * delete_requantize_headerfile
-
由 HongyuJia 提交于
-
由 Paulina Gacek 提交于
* op migrated, Copy(OneDNNContext, ...) added * mutable_data & op registration in fluid removed * refactoring * OneDNNGetDataType to uppercase * missing cpu check added, handler moved to .h file * name changed to transpose_grad * Copy changed back to TensorCopy * Resizing corrected, Copy(OneDNNContext) removed
-
- 09 10月, 2022 4 次提交
-
-
由 zhangkaihuo 提交于
-
由 zhangkaihuo 提交于
-
由 Haohongxiang 提交于
-
由 ronnywang 提交于
-
- 08 10月, 2022 2 次提交
-
-
由 cifar10 提交于
-
由 mrcangye 提交于
* fix some doc bug test=document_fix * fix some docs issues, test=document_fix * beta -> \beta in softplus * threshold -> \varepsilon in softplus * parameter name * delta -> \delta in smooth_l1_loss * fix some docs test=document_fix * fix docs test=document_fix * fix docs && 增加空行 test=document_fix * Update python/paddle/nn/functional/activation.py, test=document_fix * Update python/paddle/nn/layer/activation.py, test=document_fix Co-authored-by: NSigureMo <sigure.qaq@gmail.com>
-
- 03 10月, 2022 2 次提交
-
-
由 jakpiase 提交于
* added transpose * CI fix * fix for transpose * fix after review
-
由 Jacek Czaja 提交于
* - some more MD changes * - lint * - compilation fixes * - compilation fixes * - lint * - fix
-
- 30 9月, 2022 8 次提交
-
-
由 HongyuJia 提交于
-
由 HongyuJia 提交于
-
由 Chenxiao Niu 提交于
-
由 光明和真理 提交于
-
由 HongyuJia 提交于
* opt GetExpectedKernelType code of conv_transpose_op * fix if error
-
由 HongyuJia 提交于
-
由 HongyuJia 提交于
-
由 sneaxiy 提交于
* support pure bfloat16 * support bf16 linear * update PR to pass CI * tiny fix where_grad_kernel.cu * add bfloat16 to selu_grad to pass CI * fix selu grad compilation error
-
- 29 9月, 2022 2 次提交
-
-
由 carryyu 提交于
* fix P40 topk: Make the optimized topk compatible with P40. * fix P40 topk: Make the optimized topk compatible with P40. * fix P40 topk: Make the optimized topk compatible with P40.
-
由 光明和真理 提交于
Co-authored-by: Nliupeiyu <liupeiyu@cambricon.com>
-
- 28 9月, 2022 5 次提交
-
-
由 Chen Weihang 提交于
* remove needless using tensor * remove needless using tensor * resolve conflict * replace tensor using * fix format error * revert needless changing * fix rocm and npu compile error * fix cinn compile error * fix format error * fix mkldnn format error * fix mkldnn format error * fix cinn compile error * fix cinn compile error * fix cinn compile error * resolve conflict
-
由 Jacek Czaja 提交于
* added fc int8 tests * CI fix * added skipping UTs for GPUs * fixes for CI * added support for residual connections inside fc * fix for quant int8 bias * - lint Co-authored-by: Njakpiase <jakpia21@gmail.com>
-
由 limingshu 提交于
-
由 Sławomir Siwek 提交于
* Relu6 * remove fluid handler * add individual kernel signature * coding style * replace bounded_relu with clip * whitespace * code style
-
由 zyfncg 提交于
* fix clip_extra code in remove_training_info * revert rnn opmaker clear
-