提交 · c3df3f2d956dbd640e33eb9562e3b6cdfca6221b · PaddlePaddle / Paddle-Lite

11 2月, 2020 1 次提交

Optimize the compute implementation of several operators. (#2843) · 53604bac

由 Yiqun Liu 提交于 2月 11, 2020

* Optimize the transform from Paddle' Tensor to EigenVector, avoiding defining multiple DDim.

* Optimize the compute implementation of several operators.
test=develop

53604bac

06 2月, 2020 1 次提交
- L
  
  Save the workspace. · 118c1441
  由 Liu Yiqun 提交于 2月 06, 2020
  
  118c1441
24 12月, 2019 1 次提交

[LITE][NPU][XPU] Support multiple types for XPU and NPU op bridges (#2646) · 05da0c72

由 hong19860320 提交于 12月 24, 2019

* Support multiple types for XPU and NPU op bridges

* Add lookup_table, gather, slice, stack and scale op bridges for supporting BERT

* Fix the definition of lookup_table kernel for X86

05da0c72

25 11月, 2019 1 次提交
- W
  update x86 op and kernel to run content-dnn model test=develop (#2481) · a1a41934
  由 Wilber 提交于 11月 25, 2019
```
* update x86 op and kernel to run content-dnn model test=develop
```
  a1a41934
14 11月, 2019 1 次提交
- P
  Update lookup_table op on arm x86，add lookup_table_v2_op (#2405) · 94731268
  由 Pei Yang 提交于 11月 14, 2019
```
* update lookup_table arm x86, test=develop

* add lookup_table_v2_op for compatibility, test=develop
```
  94731268
05 11月, 2019 1 次提交
- L
  fix StepRNN model run related bugs (#2300) · 7f5c0ca1
  由 lijianshe02 提交于 11月 05, 2019
```
* fix step rnn model run bugs test=develop
```
  7f5c0ca1