File size: 8,384 Bytes
b213d84 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 |
// Copyright (c) Facebook, Inc. and its affiliates.
#pragma once
#include <torch/types.h>
namespace detectron2 {
#if defined(WITH_CUDA) || defined(WITH_HIP)
int deform_conv_forward_cuda(
at::Tensor input,
at::Tensor weight,
at::Tensor offset,
at::Tensor output,
at::Tensor columns,
at::Tensor ones,
int kW,
int kH,
int dW,
int dH,
int padW,
int padH,
int dilationW,
int dilationH,
int group,
int deformable_group,
int im2col_step);
int deform_conv_backward_input_cuda(
at::Tensor input,
at::Tensor offset,
at::Tensor gradOutput,
at::Tensor gradInput,
at::Tensor gradOffset,
at::Tensor weight,
at::Tensor columns,
int kW,
int kH,
int dW,
int dH,
int padW,
int padH,
int dilationW,
int dilationH,
int group,
int deformable_group,
int im2col_step);
int deform_conv_backward_parameters_cuda(
at::Tensor input,
at::Tensor offset,
at::Tensor gradOutput,
at::Tensor gradWeight, // at::Tensor gradBias,
at::Tensor columns,
at::Tensor ones,
int kW,
int kH,
int dW,
int dH,
int padW,
int padH,
int dilationW,
int dilationH,
int group,
int deformable_group,
float scale,
int im2col_step);
void modulated_deform_conv_cuda_forward(
at::Tensor input,
at::Tensor weight,
at::Tensor bias,
at::Tensor ones,
at::Tensor offset,
at::Tensor mask,
at::Tensor output,
at::Tensor columns,
int kernel_h,
int kernel_w,
const int stride_h,
const int stride_w,
const int pad_h,
const int pad_w,
const int dilation_h,
const int dilation_w,
const int group,
const int deformable_group,
const bool with_bias);
void modulated_deform_conv_cuda_backward(
at::Tensor input,
at::Tensor weight,
at::Tensor bias,
at::Tensor ones,
at::Tensor offset,
at::Tensor mask,
at::Tensor columns,
at::Tensor grad_input,
at::Tensor grad_weight,
at::Tensor grad_bias,
at::Tensor grad_offset,
at::Tensor grad_mask,
at::Tensor grad_output,
int kernel_h,
int kernel_w,
int stride_h,
int stride_w,
int pad_h,
int pad_w,
int dilation_h,
int dilation_w,
int group,
int deformable_group,
const bool with_bias);
#endif
inline int deform_conv_forward(
at::Tensor input,
at::Tensor weight,
at::Tensor offset,
at::Tensor output,
at::Tensor columns,
at::Tensor ones,
int kW,
int kH,
int dW,
int dH,
int padW,
int padH,
int dilationW,
int dilationH,
int group,
int deformable_group,
int im2col_step) {
if (input.is_cuda()) {
#if defined(WITH_CUDA) || defined(WITH_HIP)
TORCH_CHECK(weight.is_cuda(), "weight tensor is not on GPU!");
TORCH_CHECK(offset.is_cuda(), "offset tensor is not on GPU!");
return deform_conv_forward_cuda(
input,
weight,
offset,
output,
columns,
ones,
kW,
kH,
dW,
dH,
padW,
padH,
dilationW,
dilationH,
group,
deformable_group,
im2col_step);
#else
AT_ERROR("Detectron2 is not compiled with GPU support!");
#endif
}
AT_ERROR("This operator is not implemented on CPU");
}
inline int deform_conv_backward_input(
at::Tensor input,
at::Tensor offset,
at::Tensor gradOutput,
at::Tensor gradInput,
at::Tensor gradOffset,
at::Tensor weight,
at::Tensor columns,
int kW,
int kH,
int dW,
int dH,
int padW,
int padH,
int dilationW,
int dilationH,
int group,
int deformable_group,
int im2col_step) {
if (gradOutput.is_cuda()) {
#if defined(WITH_CUDA) || defined(WITH_HIP)
TORCH_CHECK(input.is_cuda(), "input tensor is not on GPU!");
TORCH_CHECK(weight.is_cuda(), "weight tensor is not on GPU!");
TORCH_CHECK(offset.is_cuda(), "offset tensor is not on GPU!");
return deform_conv_backward_input_cuda(
input,
offset,
gradOutput,
gradInput,
gradOffset,
weight,
columns,
kW,
kH,
dW,
dH,
padW,
padH,
dilationW,
dilationH,
group,
deformable_group,
im2col_step);
#else
AT_ERROR("Detectron2 is not compiled with GPU support!");
#endif
}
AT_ERROR("This operator is not implemented on CPU");
}
inline int deform_conv_backward_filter(
at::Tensor input,
at::Tensor offset,
at::Tensor gradOutput,
at::Tensor gradWeight, // at::Tensor gradBias,
at::Tensor columns,
at::Tensor ones,
int kW,
int kH,
int dW,
int dH,
int padW,
int padH,
int dilationW,
int dilationH,
int group,
int deformable_group,
float scale,
int im2col_step) {
if (gradOutput.is_cuda()) {
#if defined(WITH_CUDA) || defined(WITH_HIP)
TORCH_CHECK(input.is_cuda(), "input tensor is not on GPU!");
TORCH_CHECK(offset.is_cuda(), "offset tensor is not on GPU!");
return deform_conv_backward_parameters_cuda(
input,
offset,
gradOutput,
gradWeight,
columns,
ones,
kW,
kH,
dW,
dH,
padW,
padH,
dilationW,
dilationH,
group,
deformable_group,
scale,
im2col_step);
#else
AT_ERROR("Detectron2 is not compiled with GPU support!");
#endif
}
AT_ERROR("This operator is not implemented on CPU");
}
inline void modulated_deform_conv_forward(
at::Tensor input,
at::Tensor weight,
at::Tensor bias,
at::Tensor ones,
at::Tensor offset,
at::Tensor mask,
at::Tensor output,
at::Tensor columns,
int kernel_h,
int kernel_w,
const int stride_h,
const int stride_w,
const int pad_h,
const int pad_w,
const int dilation_h,
const int dilation_w,
const int group,
const int deformable_group,
const bool with_bias) {
if (input.is_cuda()) {
#if defined(WITH_CUDA) || defined(WITH_HIP)
TORCH_CHECK(weight.is_cuda(), "weight tensor is not on GPU!");
TORCH_CHECK(bias.is_cuda(), "bias tensor is not on GPU!");
TORCH_CHECK(offset.is_cuda(), "offset tensor is not on GPU!");
return modulated_deform_conv_cuda_forward(
input,
weight,
bias,
ones,
offset,
mask,
output,
columns,
kernel_h,
kernel_w,
stride_h,
stride_w,
pad_h,
pad_w,
dilation_h,
dilation_w,
group,
deformable_group,
with_bias);
#else
AT_ERROR("Detectron2 is not compiled with GPU support!");
#endif
}
AT_ERROR("This operator is not implemented on CPU");
}
inline void modulated_deform_conv_backward(
at::Tensor input,
at::Tensor weight,
at::Tensor bias,
at::Tensor ones,
at::Tensor offset,
at::Tensor mask,
at::Tensor columns,
at::Tensor grad_input,
at::Tensor grad_weight,
at::Tensor grad_bias,
at::Tensor grad_offset,
at::Tensor grad_mask,
at::Tensor grad_output,
int kernel_h,
int kernel_w,
int stride_h,
int stride_w,
int pad_h,
int pad_w,
int dilation_h,
int dilation_w,
int group,
int deformable_group,
const bool with_bias) {
if (grad_output.is_cuda()) {
#if defined(WITH_CUDA) || defined(WITH_HIP)
TORCH_CHECK(input.is_cuda(), "input tensor is not on GPU!");
TORCH_CHECK(weight.is_cuda(), "weight tensor is not on GPU!");
TORCH_CHECK(bias.is_cuda(), "bias tensor is not on GPU!");
TORCH_CHECK(offset.is_cuda(), "offset tensor is not on GPU!");
return modulated_deform_conv_cuda_backward(
input,
weight,
bias,
ones,
offset,
mask,
columns,
grad_input,
grad_weight,
grad_bias,
grad_offset,
grad_mask,
grad_output,
kernel_h,
kernel_w,
stride_h,
stride_w,
pad_h,
pad_w,
dilation_h,
dilation_w,
group,
deformable_group,
with_bias);
#else
AT_ERROR("Detectron2 is not compiled with GPU support!");
#endif
}
AT_ERROR("This operator is not implemented on CPU");
}
} // namespace detectron2
|