Unverified Commit c959dab8 authored by Krishna Murthy's avatar Krishna Murthy Committed by GitHub
Browse files

Make compatible with pytorch 1.11 and newer; bugfix (#418)



* Bugfix: missing include common.h
Signed-off-by: default avatarKrishna Murthy <krrish94@gmail.com>

* Make compatible with pytorch 1.11 and newer -- THCudaCheck() deprecated
Signed-off-by: default avatarKrishna Murthy <krrish94@gmail.com>

---------
Signed-off-by: default avatarKrishna Murthy <krrish94@gmail.com>
parent 331ecdd5
......@@ -19,8 +19,6 @@
#include <ATen/ATen.h>
#include <ATen/cuda/CUDAContext.h>
#include <THC/THCNumerics.cuh>
#include <THC/THC.h>
#include <cuda.h>
......@@ -378,14 +376,14 @@ std::vector<at::Tensor> box_encoder(const int N_img,
printf("allocating %lu bytes for output labels\n", N_img*M*sizeof(long));
#endif
at::Tensor labels_out = at::empty({N_img * M}, labels_input.options());
THCudaCheck(cudaGetLastError());
C10_CUDA_CHECK(cudaGetLastError());
// copy default boxes to outputs
#ifdef DEBUG
printf("allocating %lu bytes for output bboxes\n", N_img*M*4*sizeof(float));
#endif
at::Tensor bbox_out = dbox.repeat({N_img, 1});
THCudaCheck(cudaGetLastError());
C10_CUDA_CHECK(cudaGetLastError());
// need to allocate some workspace
#ifdef DEBUG
......@@ -393,7 +391,7 @@ std::vector<at::Tensor> box_encoder(const int N_img,
#endif
// at::Tensor workspace = at::CUDA(at::kByte).zeros({8 * M * N_img});
at::Tensor workspace = at::zeros({8 * M * N_img}, at::CUDA(at::kByte));
THCudaCheck(cudaGetLastError());
C10_CUDA_CHECK(cudaGetLastError());
// Encode the inputs
const int THREADS_PER_BLOCK = 256;
......@@ -408,7 +406,7 @@ std::vector<at::Tensor> box_encoder(const int N_img,
(float4*)bbox_out.data_ptr<float>(),
labels_out.data_ptr<long>());
THCudaCheck(cudaGetLastError());
C10_CUDA_CHECK(cudaGetLastError());
return {bbox_out, labels_out};
}
......@@ -435,7 +433,7 @@ at::Tensor calc_ious(const int N_img,
(float4*)boxes2.data_ptr<float>(),
ious.data_ptr<float>());
THCudaCheck(cudaGetLastError());
C10_CUDA_CHECK(cudaGetLastError());
return ious;
}
......@@ -569,7 +567,7 @@ std::vector<at::Tensor> random_horiz_flip(
flip.data_ptr<float>(),
tmp_img.data_ptr<scalar_t>(),
nhwc);
THCudaCheck(cudaGetLastError());
C10_CUDA_CHECK(cudaGetLastError());
});
// copy tmp_img -> img
......
......@@ -10,6 +10,8 @@
#include <ATen/cuda/detail/IndexUtils.cuh>
#include <ATen/cuda/detail/KernelUtils.h>
#include "common.h"
template <typename dest_t, typename src_t>
static inline dest_t safe_downcast(src_t v)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment