Block or Report
Block or report jiweibo
Contact GitHub support about this user’s behavior. Learn more about reporting abuse.
Report abusePinned
-
-
Paddle Public
Forked from PaddlePaddle/Paddle
PArallel Distributed Deep LEarning (『飞桨』核心框架,高性能单机、分布式训练和跨平台部署)
Python
-
tensorflow Public
Forked from tensorflow/tensorflow
An Open Source Machine Learning Framework for Everyone
C++
-
Paddle-Lite Public
Forked from PaddlePaddle/Paddle-Lite
Multi-platform high performance deep learning inference engine (『飞桨』多平台高性能深度学习预测引擎)
C++
-
pytorch Public
Forked from pytorch/pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration
C++
-
591 contributions in the last year
Activity overview
Contribution activity
March 2022
Created 14 commits in 4 repositories
Created a pull request in PaddlePaddle/Paddle that received 2 comments
Add mlir trt engine type.
PR types Others
PR changes Others
Describe
Rename TRTEngine to TrtEngine
Add mlir trt dialect engine type. !trt.engine
Opened 4 other pull requests in 1 repository
PaddlePaddle/Paddle
1
open
3
merged
Reviewed 18 pull requests in 1 repository
PaddlePaddle/Paddle
18 pull requests
- [Cmake]fix inference_lib.cmake
- Refine to_tensor for eager mode and support gpu_pinned
- add the map for dense tensor, test=develop
- [Phi] Add phi device context pool
- [infrt]Refine phi dialect
- [infrt] add parse for infrt.dense_tensor_type.
- support gpu mixed precision inference
- Trt engine.
- adjusts the mlir attrs order, test=develop
- fix gpu callback
- Add trt execute
- fix_conv2d_trt_convert_test_case
- EmbEltwiseLayernorm fix
- [CustomRuntime] migrate CustomRuntime into phi
- fix_trt_engine_op_bug
- Inference add ONNXRuntime back-end
- [Eager] open eager when WITH_PYTHON
- add share external data interface