From c9d2046f7688c5c23ec939bf9060780d78796b35 Mon Sep 17 00:00:00 2001
From: jerrywgz <jerrywgz@126.com>
Date: Tue, 16 Oct 2018 03:02:45 +0000
Subject: [PATCH] roi_align for gpu

---
 paddle/fluid/operators/roi_align_op.cu | 38 ++++++++++++++++++++++++++
 1 file changed, 38 insertions(+)
 create mode 100644 paddle/fluid/operators/roi_align_op.cu

diff --git a/paddle/fluid/operators/roi_align_op.cu b/paddle/fluid/operators/roi_align_op.cu
new file mode 100644
index 0000000000..a35113e5f9
--- /dev/null
+++ b/paddle/fluid/operators/roi_align_op.cu
@@ -0,0 +1,38 @@
+/* Copyright (c) 2016 PaddlePaddle Authors. All Rights Reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License");
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+    http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software
+distributed under the License is distributed on an "AS IS" BASIS,
+WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+See the License for the specific language governing permissions and
+limitations under the License. */
+
+#include "paddle/fluid/operators/roi_align_op.h"
+#include "paddle/fluid/platform/cuda_primitives.h"
+
+namespace paddle {
+namespace operators {
+
+using Tensor = framework::Tensor;
+using LoDTensor = framework::LoDTensor;
+
+static constexpr int kNumCUDAThreads = 512;
+static constexpr int kNumMaxinumNumBlocks = 4096;
+
+}  // namespace operators
+}  // namespace paddle
+
+namespace ops = paddle::operators;
+REGISTER_OP_CUDA_KERNEL(
+    roi_align,
+    ops::GPUROIAlignOpKernel<paddle::platform::CUDADeviceContext, float>,
+    ops::GPUROIAlignOpKernel<paddle::platform::CUDADeviceContext, double>);
+REGISTER_OP_CUDA_KERNEL(
+    roi_align_grad,
+    ops::GPUROIAlignGradOpKernel<paddle::platform::CUDADeviceContext, float>,
+    ops::GPUROIAlignGradOpKernel<paddle::platform::CUDADeviceContext, double>);
-- 
GitLab