Skip to content

P
Paddle

BaiXuePrincess / Paddle
与 Fork 源项目一致

Fork自 PaddlePaddle / Paddle

通知 1
Star 1
Fork 0
P
Paddle
未验证 提交 7b627dd8 编写于 作者: H hong 提交者： GitHub
浏览文件
操作

fix depthwise dnn bug (#41666)

上级 3749198e
显示空白变更内容
内联 并排
Showing with 2 addition and 0 deletion
paddle/phi/kernels/gpudnn/conv_grad_kernel.cu
浏览文件 @ 7b627dd8
...@@ -641,6 +641,7 @@ void DepthwiseConvCudnnGradKernel(const Context& dev_ctx, ...@@ -641,6 +641,7 @@ void DepthwiseConvCudnnGradKernel(const Context& dev_ctx,
bool use_addto, bool use_addto,
int workspace_size_MB, int workspace_size_MB,
bool exhaustive_search, bool exhaustive_search,
bool fuse_relu,
DenseTensor* input_grad, DenseTensor* input_grad,
DenseTensor* filter_grad) { DenseTensor* filter_grad) {
ConvCudnnGradKernel<T>(dev_ctx, ConvCudnnGradKernel<T>(dev_ctx,
......
paddle/phi/kernels/gpudnn/conv_kernel.cu
浏览文件 @ 7b627dd8
...@@ -429,6 +429,7 @@ void DepthwiseConvCudnnKernel(const Context& dev_ctx, ...@@ -429,6 +429,7 @@ void DepthwiseConvCudnnKernel(const Context& dev_ctx,
bool use_addto, bool use_addto,
int workspace_size_MB, int workspace_size_MB,
bool exhaustive_search, bool exhaustive_search,
bool fuse_relu,
DenseTensor* out) { DenseTensor* out) {
ConvCudnnKernel<T>(dev_ctx, ConvCudnnKernel<T>(dev_ctx,
input, input,
......
Markdown is supported
0% 或 .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑！
想要评论请 注册 或