You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

Qi Li 334296306c [ROCM] update fluid platform for rocm39 (part4), test=develop (#30936)		4 years ago
..
CMakeLists.txt	[ROCM] update fluid platform for rocm35 (part1), test=develop (#30639)	4 years ago
cublas.cc	fix cublas warp error	6 years ago
cublas.h	[Complex] Add support for complex grad accumulated (#29889)	4 years ago
cuda_driver.cc	remove WITH_DSO compile option (#25444)	5 years ago
cuda_driver.h	use iwyu clean include (#27267)	4 years ago
cudnn.cc	[bug fix]:Memory increases after adapting the cudnn version to cudnn8 (#27436)	4 years ago
cudnn.h	tune backward filter algorithm for float16 (#27529)	4 years ago
cupti.cc	use iwyu clean include (#27267)	4 years ago
cupti.h	remove WITH_DSO compile option (#25444)	5 years ago
cupti_lib_path.h.in	Fix the compilation on CUDA 9.1/GCC 5.3	7 years ago
curand.cc	Fix the grammar in copyright. (#8403)	7 years ago
curand.h	use iwyu clean include (#27267)	4 years ago
cusolver.cc	Fix cusolver loader for Windows (#24157)	5 years ago
cusolver.h	use iwyu clean include (#27267)	4 years ago
dynamic_loader.cc	[ROCM] update fluid platform for rocm39 (part4), test=develop (#30936)	4 years ago
dynamic_loader.h	use nvtx push pop in timeline (#30567)	4 years ago
hiprand.cc	添加rocm平台支持代码 (#29342)	4 years ago
hiprand.h	添加rocm平台支持代码 (#29342)	4 years ago
hiprtc.cc	添加rocm平台支持代码 (#29342)	4 years ago
hiprtc.h	[ROCM] update fluid platform for rocm39 (part4), test=develop (#30936)	4 years ago
miopen.cc	添加rocm平台支持代码 (#29342)	4 years ago
miopen.h	[ROCM] update fluid platform for rocm39 (part4), test=develop (#30936)	4 years ago
mklml.cc	Fix the definition issue when used mkl_scsrmm and mkl_dcsrmm functions. (#19774)	5 years ago
mklml.h	add complex64 and complex128 type; add +-*/@ and slice opreator for c… (#29199)	4 years ago
nccl.cc	bug fix, test=develop (#28674)	4 years ago
nccl.h	add ncclSend and ncclRecv (#27621)	4 years ago
nvrtc.cc	remove WITH_DSO compile option (#25444)	5 years ago
nvrtc.h	use iwyu clean include (#27267)	4 years ago
nvtx.cc	use nvtx push pop in timeline (#30567)	4 years ago
nvtx.h	use nvtx push pop in timeline (#30567)	4 years ago
rccl.cc	[ROCM] update fluid platform for rocm35 (part1), test=develop (#30639)	4 years ago
rccl.h	[ROCM] update fluid platform for rocm35 (part1), test=develop (#30639)	4 years ago
rocblas.cc	添加rocm平台支持代码 (#29342)	4 years ago
rocblas.h	[ROCM] update fluid platform for rocm35 (part1), test=develop (#30639)	4 years ago
rocm_driver.cc	添加rocm平台支持代码 (#29342)	4 years ago
rocm_driver.h	[ROCM] update fluid platform for rocm39 (part4), test=develop (#30936)	4 years ago
tensorrt.cc	change avg pooling and global pooling to trt layer in dynamic shape mode (#28702)	4 years ago
tensorrt.h	use iwyu clean include second time, test=develop (#30829)	4 years ago
warpctc.cc	Fix the grammar in copyright. (#8403)	7 years ago
warpctc.h	add support to float64 input of warpctc op. (#27399)	4 years ago