File size: 59,920 Bytes
0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab fc2c13a 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 97d74ab 0ab32f0 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 |
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Converting PyTorch to ONNX"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"True\n"
]
}
],
"source": [
"import torch\n",
"from torch import nn\n",
"from torch.nn import functional as F\n",
"\n",
"print(torch.cuda.is_available())"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Defining the model"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"class BasicBlock(nn.Module):\n",
"\n",
" def __init__(self, in_channels, out_channels, stride= 1):\n",
" super().__init__()\n",
" self.conv1 = nn.Conv2d(in_channels, out_channels, 3, stride= stride, padding= 1)\n",
" self.conv2 = nn.Conv2d(out_channels, out_channels, 3, stride= 1, padding= 1)\n",
" self.bn1 = nn.BatchNorm2d(out_channels)\n",
" self.bn2 = nn.BatchNorm2d(out_channels)\n",
" self.relu = nn.ReLU(inplace= True)\n",
"\n",
" self.downsample = nn.Sequential()\n",
" \n",
" if stride != 1:\n",
" self.downsample = nn.Sequential(\n",
" nn.Conv2d(in_channels, out_channels, 1, stride),\n",
" nn.BatchNorm2d(out_channels))\n",
"\n",
" def forward(self, x):\n",
" out = self.conv1(x)\n",
" out = self.bn1(out)\n",
" out = self.relu(out)\n",
"\n",
" out = self.conv2(out)\n",
" out = self.bn2(out)\n",
"\n",
" out += self.downsample(x)\n",
"\n",
" out = self.relu(out)\n",
"\n",
" return out\n",
"\n",
"class ResNet34(nn.Module):\n",
" def __init__(self, num_classes) -> None:\n",
" super().__init__()\n",
"\n",
" self.conv1 = nn.Sequential(nn.Conv2d(3, 64, 7, stride= 2, padding= 3),\n",
" nn.BatchNorm2d(64), nn.MaxPool2d(3, stride= 2, padding= 1), nn.ReLU(inplace= True))\n",
"\n",
" self.layer0 = self._make_layer(64, 64, 3, 1)\n",
" self.layer1 = self._make_layer(64, 128, 4, 2)\n",
" self.layer2 = self._make_layer(128, 256, 6, 2)\n",
" self.layer3 = self._make_layer(256, 512, 3, 2)\n",
"\n",
" self.avg_pool = nn.AvgPool2d(7)\n",
" self.fc = nn.Linear(512, num_classes)\n",
"\n",
" def _make_layer(self, in_channels, out_channels, num_blocks, stride):\n",
" \n",
" layers = []\n",
" layers.append(BasicBlock(in_channels, out_channels, stride))\n",
"\n",
" for i in range(num_blocks - 1):\n",
" layers.append(BasicBlock(out_channels, out_channels, 1))\n",
"\n",
" return nn.Sequential(*layers)\n",
"\n",
" def forward(self, x):\n",
" out = self.conv1(x)\n",
" \n",
" out = self.layer0(out)\n",
" out = self.layer1(out)\n",
" out = self.layer2(out)\n",
" out = self.layer3(out)\n",
"\n",
" out = self.avg_pool(out)\n",
" out = out.view(out.size(0), -1)\n",
" out = self.fc(out)\n",
"\n",
" out = F.log_softmax(out, dim= 1)\n",
" return out"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Loading the model"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"model = ResNet34(450)\n",
"model.load_state_dict(torch.load('../model/birds_resnet34_model.pth'))\n",
"\n",
"dummy_input = torch.randn(1, 3, 224, 224)\n",
"\n",
"model.eval()\n",
"torch_out = model(dummy_input)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Converting to ONNX"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Exported graph: graph(%input : Float(*, 3, 224, 224, strides=[150528, 50176, 224, 1], requires_grad=0, device=cpu),\n",
" %fc.weight : Float(450, 512, strides=[512, 1], requires_grad=1, device=cpu),\n",
" %fc.bias : Float(450, strides=[1], requires_grad=1, device=cpu),\n",
" %onnx::Conv_393 : Float(64, 3, 7, 7, strides=[147, 49, 7, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_394 : Float(64, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_396 : Float(64, 64, 3, 3, strides=[576, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_397 : Float(64, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_399 : Float(64, 64, 3, 3, strides=[576, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_400 : Float(64, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_402 : Float(64, 64, 3, 3, strides=[576, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_403 : Float(64, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_405 : Float(64, 64, 3, 3, strides=[576, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_406 : Float(64, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_408 : Float(64, 64, 3, 3, strides=[576, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_409 : Float(64, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_411 : Float(64, 64, 3, 3, strides=[576, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_412 : Float(64, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_414 : Float(128, 64, 3, 3, strides=[576, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_415 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_417 : Float(128, 128, 3, 3, strides=[1152, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_418 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_420 : Float(128, 64, 1, 1, strides=[64, 1, 1, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_421 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_423 : Float(128, 128, 3, 3, strides=[1152, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_424 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_426 : Float(128, 128, 3, 3, strides=[1152, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_427 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_429 : Float(128, 128, 3, 3, strides=[1152, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_430 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_432 : Float(128, 128, 3, 3, strides=[1152, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_433 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_435 : Float(128, 128, 3, 3, strides=[1152, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_436 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_438 : Float(128, 128, 3, 3, strides=[1152, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_439 : Float(128, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_441 : Float(256, 128, 3, 3, strides=[1152, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_442 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_444 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_445 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_447 : Float(256, 128, 1, 1, strides=[128, 1, 1, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_448 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_450 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_451 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_453 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_454 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_456 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_457 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_459 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_460 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_462 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_463 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_465 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_466 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_468 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_469 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_471 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_472 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_474 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_475 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_477 : Float(256, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_478 : Float(256, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_480 : Float(512, 256, 3, 3, strides=[2304, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_481 : Float(512, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_483 : Float(512, 512, 3, 3, strides=[4608, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_484 : Float(512, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_486 : Float(512, 256, 1, 1, strides=[256, 1, 1, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_487 : Float(512, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_489 : Float(512, 512, 3, 3, strides=[4608, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_490 : Float(512, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_492 : Float(512, 512, 3, 3, strides=[4608, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_493 : Float(512, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_495 : Float(512, 512, 3, 3, strides=[4608, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_496 : Float(512, strides=[1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_498 : Float(512, 512, 3, 3, strides=[4608, 9, 3, 1], requires_grad=0, device=cpu),\n",
" %onnx::Conv_499 : Float(512, strides=[1], requires_grad=0, device=cpu)):\n",
" %/conv1/conv1.0/Conv_output_0 : Float(*, 64, 112, 112, strides=[802816, 12544, 112, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[7, 7], pads=[3, 3, 3, 3], strides=[2, 2], onnx_name=\"/conv1/conv1.0/Conv\"](%input, %onnx::Conv_393, %onnx::Conv_394), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::conv1/torch.nn.modules.conv.Conv2d::conv1.0 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/conv1/conv1.2/MaxPool_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::MaxPool[ceil_mode=0, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[2, 2], onnx_name=\"/conv1/conv1.2/MaxPool\"](%/conv1/conv1.0/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::conv1/torch.nn.modules.pooling.MaxPool2d::conv1.2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:780:0\n",
" %/conv1/conv1.3/Relu_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/conv1/conv1.3/Relu\"](%/conv1/conv1.2/MaxPool_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::conv1/torch.nn.modules.activation.ReLU::conv1.3 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer0/layer0.0/conv1/Conv_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer0/layer0.0/conv1/Conv\"](%/conv1/conv1.3/Relu_output_0, %onnx::Conv_396, %onnx::Conv_397), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.0/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer0/layer0.0/relu/Relu_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer0/layer0.0/relu/Relu\"](%/layer0/layer0.0/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.0/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer0/layer0.0/conv2/Conv_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer0/layer0.0/conv2/Conv\"](%/layer0/layer0.0/relu/Relu_output_0, %onnx::Conv_399, %onnx::Conv_400), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.0/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer0/layer0.0/Add_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer0/layer0.0/Add\"](%/layer0/layer0.0/conv2/Conv_output_0, %/conv1/conv1.3/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.0 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer0/layer0.0/relu_1/Relu_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer0/layer0.0/relu_1/Relu\"](%/layer0/layer0.0/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.0/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer0/layer0.1/conv1/Conv_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer0/layer0.1/conv1/Conv\"](%/layer0/layer0.0/relu_1/Relu_output_0, %onnx::Conv_402, %onnx::Conv_403), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.1/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer0/layer0.1/relu/Relu_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer0/layer0.1/relu/Relu\"](%/layer0/layer0.1/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.1/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer0/layer0.1/conv2/Conv_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer0/layer0.1/conv2/Conv\"](%/layer0/layer0.1/relu/Relu_output_0, %onnx::Conv_405, %onnx::Conv_406), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.1/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer0/layer0.1/Add_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer0/layer0.1/Add\"](%/layer0/layer0.1/conv2/Conv_output_0, %/layer0/layer0.0/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.1 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer0/layer0.1/relu_1/Relu_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer0/layer0.1/relu_1/Relu\"](%/layer0/layer0.1/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.1/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer0/layer0.2/conv1/Conv_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer0/layer0.2/conv1/Conv\"](%/layer0/layer0.1/relu_1/Relu_output_0, %onnx::Conv_408, %onnx::Conv_409), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.2/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer0/layer0.2/relu/Relu_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer0/layer0.2/relu/Relu\"](%/layer0/layer0.2/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.2/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer0/layer0.2/conv2/Conv_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer0/layer0.2/conv2/Conv\"](%/layer0/layer0.2/relu/Relu_output_0, %onnx::Conv_411, %onnx::Conv_412), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.2/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer0/layer0.2/Add_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer0/layer0.2/Add\"](%/layer0/layer0.2/conv2/Conv_output_0, %/layer0/layer0.1/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.2 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer0/layer0.2/relu_1/Relu_output_0 : Float(*, 64, 56, 56, strides=[200704, 3136, 56, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer0/layer0.2/relu_1/Relu\"](%/layer0/layer0.2/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer0/__main__.BasicBlock::layer0.2/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer1/layer1.0/conv1/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[2, 2], onnx_name=\"/layer1/layer1.0/conv1/Conv\"](%/layer0/layer0.2/relu_1/Relu_output_0, %onnx::Conv_414, %onnx::Conv_415), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.0/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.0/relu/Relu_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer1/layer1.0/relu/Relu\"](%/layer1/layer1.0/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.0/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer1/layer1.0/conv2/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer1/layer1.0/conv2/Conv\"](%/layer1/layer1.0/relu/Relu_output_0, %onnx::Conv_417, %onnx::Conv_418), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.0/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.0/downsample/downsample.0/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[1, 1], pads=[0, 0, 0, 0], strides=[2, 2], onnx_name=\"/layer1/layer1.0/downsample/downsample.0/Conv\"](%/layer0/layer0.2/relu_1/Relu_output_0, %onnx::Conv_420, %onnx::Conv_421), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.0/torch.nn.modules.container.Sequential::downsample/torch.nn.modules.conv.Conv2d::downsample.0 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.0/Add_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer1/layer1.0/Add\"](%/layer1/layer1.0/conv2/Conv_output_0, %/layer1/layer1.0/downsample/downsample.0/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.0 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer1/layer1.0/relu_1/Relu_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer1/layer1.0/relu_1/Relu\"](%/layer1/layer1.0/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.0/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer1/layer1.1/conv1/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer1/layer1.1/conv1/Conv\"](%/layer1/layer1.0/relu_1/Relu_output_0, %onnx::Conv_423, %onnx::Conv_424), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.1/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.1/relu/Relu_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer1/layer1.1/relu/Relu\"](%/layer1/layer1.1/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.1/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer1/layer1.1/conv2/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer1/layer1.1/conv2/Conv\"](%/layer1/layer1.1/relu/Relu_output_0, %onnx::Conv_426, %onnx::Conv_427), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.1/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.1/Add_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer1/layer1.1/Add\"](%/layer1/layer1.1/conv2/Conv_output_0, %/layer1/layer1.0/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.1 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer1/layer1.1/relu_1/Relu_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer1/layer1.1/relu_1/Relu\"](%/layer1/layer1.1/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.1/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer1/layer1.2/conv1/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer1/layer1.2/conv1/Conv\"](%/layer1/layer1.1/relu_1/Relu_output_0, %onnx::Conv_429, %onnx::Conv_430), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.2/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.2/relu/Relu_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer1/layer1.2/relu/Relu\"](%/layer1/layer1.2/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.2/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer1/layer1.2/conv2/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer1/layer1.2/conv2/Conv\"](%/layer1/layer1.2/relu/Relu_output_0, %onnx::Conv_432, %onnx::Conv_433), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.2/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.2/Add_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer1/layer1.2/Add\"](%/layer1/layer1.2/conv2/Conv_output_0, %/layer1/layer1.1/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.2 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer1/layer1.2/relu_1/Relu_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer1/layer1.2/relu_1/Relu\"](%/layer1/layer1.2/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.2/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer1/layer1.3/conv1/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer1/layer1.3/conv1/Conv\"](%/layer1/layer1.2/relu_1/Relu_output_0, %onnx::Conv_435, %onnx::Conv_436), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.3/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.3/relu/Relu_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer1/layer1.3/relu/Relu\"](%/layer1/layer1.3/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.3/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer1/layer1.3/conv2/Conv_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer1/layer1.3/conv2/Conv\"](%/layer1/layer1.3/relu/Relu_output_0, %onnx::Conv_438, %onnx::Conv_439), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.3/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer1/layer1.3/Add_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer1/layer1.3/Add\"](%/layer1/layer1.3/conv2/Conv_output_0, %/layer1/layer1.2/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.3 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer1/layer1.3/relu_1/Relu_output_0 : Float(*, 128, 28, 28, strides=[100352, 784, 28, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer1/layer1.3/relu_1/Relu\"](%/layer1/layer1.3/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer1/__main__.BasicBlock::layer1.3/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.0/conv1/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[2, 2], onnx_name=\"/layer2/layer2.0/conv1/Conv\"](%/layer1/layer1.3/relu_1/Relu_output_0, %onnx::Conv_441, %onnx::Conv_442), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.0/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.0/relu/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.0/relu/Relu\"](%/layer2/layer2.0/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.0/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.0/conv2/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.0/conv2/Conv\"](%/layer2/layer2.0/relu/Relu_output_0, %onnx::Conv_444, %onnx::Conv_445), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.0/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.0/downsample/downsample.0/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[1, 1], pads=[0, 0, 0, 0], strides=[2, 2], onnx_name=\"/layer2/layer2.0/downsample/downsample.0/Conv\"](%/layer1/layer1.3/relu_1/Relu_output_0, %onnx::Conv_447, %onnx::Conv_448), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.0/torch.nn.modules.container.Sequential::downsample/torch.nn.modules.conv.Conv2d::downsample.0 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.0/Add_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer2/layer2.0/Add\"](%/layer2/layer2.0/conv2/Conv_output_0, %/layer2/layer2.0/downsample/downsample.0/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.0 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer2/layer2.0/relu_1/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.0/relu_1/Relu\"](%/layer2/layer2.0/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.0/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.1/conv1/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.1/conv1/Conv\"](%/layer2/layer2.0/relu_1/Relu_output_0, %onnx::Conv_450, %onnx::Conv_451), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.1/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.1/relu/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.1/relu/Relu\"](%/layer2/layer2.1/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.1/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.1/conv2/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.1/conv2/Conv\"](%/layer2/layer2.1/relu/Relu_output_0, %onnx::Conv_453, %onnx::Conv_454), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.1/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.1/Add_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer2/layer2.1/Add\"](%/layer2/layer2.1/conv2/Conv_output_0, %/layer2/layer2.0/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.1 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer2/layer2.1/relu_1/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.1/relu_1/Relu\"](%/layer2/layer2.1/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.1/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.2/conv1/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.2/conv1/Conv\"](%/layer2/layer2.1/relu_1/Relu_output_0, %onnx::Conv_456, %onnx::Conv_457), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.2/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.2/relu/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.2/relu/Relu\"](%/layer2/layer2.2/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.2/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.2/conv2/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.2/conv2/Conv\"](%/layer2/layer2.2/relu/Relu_output_0, %onnx::Conv_459, %onnx::Conv_460), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.2/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.2/Add_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer2/layer2.2/Add\"](%/layer2/layer2.2/conv2/Conv_output_0, %/layer2/layer2.1/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.2 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer2/layer2.2/relu_1/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.2/relu_1/Relu\"](%/layer2/layer2.2/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.2/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.3/conv1/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.3/conv1/Conv\"](%/layer2/layer2.2/relu_1/Relu_output_0, %onnx::Conv_462, %onnx::Conv_463), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.3/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.3/relu/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.3/relu/Relu\"](%/layer2/layer2.3/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.3/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.3/conv2/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.3/conv2/Conv\"](%/layer2/layer2.3/relu/Relu_output_0, %onnx::Conv_465, %onnx::Conv_466), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.3/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.3/Add_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer2/layer2.3/Add\"](%/layer2/layer2.3/conv2/Conv_output_0, %/layer2/layer2.2/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.3 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer2/layer2.3/relu_1/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.3/relu_1/Relu\"](%/layer2/layer2.3/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.3/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.4/conv1/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.4/conv1/Conv\"](%/layer2/layer2.3/relu_1/Relu_output_0, %onnx::Conv_468, %onnx::Conv_469), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.4/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.4/relu/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.4/relu/Relu\"](%/layer2/layer2.4/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.4/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.4/conv2/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.4/conv2/Conv\"](%/layer2/layer2.4/relu/Relu_output_0, %onnx::Conv_471, %onnx::Conv_472), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.4/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.4/Add_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer2/layer2.4/Add\"](%/layer2/layer2.4/conv2/Conv_output_0, %/layer2/layer2.3/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.4 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer2/layer2.4/relu_1/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.4/relu_1/Relu\"](%/layer2/layer2.4/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.4/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.5/conv1/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.5/conv1/Conv\"](%/layer2/layer2.4/relu_1/Relu_output_0, %onnx::Conv_474, %onnx::Conv_475), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.5/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.5/relu/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.5/relu/Relu\"](%/layer2/layer2.5/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.5/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer2/layer2.5/conv2/Conv_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer2/layer2.5/conv2/Conv\"](%/layer2/layer2.5/relu/Relu_output_0, %onnx::Conv_477, %onnx::Conv_478), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.5/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer2/layer2.5/Add_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer2/layer2.5/Add\"](%/layer2/layer2.5/conv2/Conv_output_0, %/layer2/layer2.4/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.5 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer2/layer2.5/relu_1/Relu_output_0 : Float(*, 256, 14, 14, strides=[50176, 196, 14, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer2/layer2.5/relu_1/Relu\"](%/layer2/layer2.5/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer2/__main__.BasicBlock::layer2.5/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer3/layer3.0/conv1/Conv_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[2, 2], onnx_name=\"/layer3/layer3.0/conv1/Conv\"](%/layer2/layer2.5/relu_1/Relu_output_0, %onnx::Conv_480, %onnx::Conv_481), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.0/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer3/layer3.0/relu/Relu_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer3/layer3.0/relu/Relu\"](%/layer3/layer3.0/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.0/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer3/layer3.0/conv2/Conv_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer3/layer3.0/conv2/Conv\"](%/layer3/layer3.0/relu/Relu_output_0, %onnx::Conv_483, %onnx::Conv_484), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.0/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer3/layer3.0/downsample/downsample.0/Conv_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[1, 1], pads=[0, 0, 0, 0], strides=[2, 2], onnx_name=\"/layer3/layer3.0/downsample/downsample.0/Conv\"](%/layer2/layer2.5/relu_1/Relu_output_0, %onnx::Conv_486, %onnx::Conv_487), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.0/torch.nn.modules.container.Sequential::downsample/torch.nn.modules.conv.Conv2d::downsample.0 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer3/layer3.0/Add_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer3/layer3.0/Add\"](%/layer3/layer3.0/conv2/Conv_output_0, %/layer3/layer3.0/downsample/downsample.0/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.0 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer3/layer3.0/relu_1/Relu_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer3/layer3.0/relu_1/Relu\"](%/layer3/layer3.0/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.0/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer3/layer3.1/conv1/Conv_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer3/layer3.1/conv1/Conv\"](%/layer3/layer3.0/relu_1/Relu_output_0, %onnx::Conv_489, %onnx::Conv_490), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.1/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer3/layer3.1/relu/Relu_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer3/layer3.1/relu/Relu\"](%/layer3/layer3.1/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.1/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer3/layer3.1/conv2/Conv_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer3/layer3.1/conv2/Conv\"](%/layer3/layer3.1/relu/Relu_output_0, %onnx::Conv_492, %onnx::Conv_493), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.1/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer3/layer3.1/Add_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer3/layer3.1/Add\"](%/layer3/layer3.1/conv2/Conv_output_0, %/layer3/layer3.0/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.1 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer3/layer3.1/relu_1/Relu_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer3/layer3.1/relu_1/Relu\"](%/layer3/layer3.1/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.1/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer3/layer3.2/conv1/Conv_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer3/layer3.2/conv1/Conv\"](%/layer3/layer3.1/relu_1/Relu_output_0, %onnx::Conv_495, %onnx::Conv_496), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.2/torch.nn.modules.conv.Conv2d::conv1 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer3/layer3.2/relu/Relu_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer3/layer3.2/relu/Relu\"](%/layer3/layer3.2/conv1/Conv_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.2/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/layer3/layer3.2/conv2/Conv_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Conv[dilations=[1, 1], group=1, kernel_shape=[3, 3], pads=[1, 1, 1, 1], strides=[1, 1], onnx_name=\"/layer3/layer3.2/conv2/Conv\"](%/layer3/layer3.2/relu/Relu_output_0, %onnx::Conv_498, %onnx::Conv_499), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.2/torch.nn.modules.conv.Conv2d::conv2 # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/conv.py:458:0\n",
" %/layer3/layer3.2/Add_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Add[onnx_name=\"/layer3/layer3.2/Add\"](%/layer3/layer3.2/conv2/Conv_output_0, %/layer3/layer3.1/relu_1/Relu_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.2 # /tmp/ipykernel_33697/1263192908.py:23:0\n",
" %/layer3/layer3.2/relu_1/Relu_output_0 : Float(*, 512, 7, 7, strides=[25088, 49, 7, 1], requires_grad=1, device=cpu) = onnx::Relu[onnx_name=\"/layer3/layer3.2/relu_1/Relu\"](%/layer3/layer3.2/Add_output_0), scope: __main__.ResNet34::/torch.nn.modules.container.Sequential::layer3/__main__.BasicBlock::layer3.2/torch.nn.modules.activation.ReLU::relu # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1453:0\n",
" %/avg_pool/Constant_output_0 : Long(8, strides=[1], device=cpu) = onnx::Constant[value= 0 0 0 0 0 0 0 0 [ CPULongType{8} ], onnx_name=\"/avg_pool/Constant\"](), scope: __main__.ResNet34::/torch.nn.modules.pooling.AvgPool2d::avg_pool # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/pooling.py:628:0\n",
" %/avg_pool/Pad_output_0 : Float(*, 512, 7, 7, device=cpu) = onnx::Pad[mode=\"constant\", onnx_name=\"/avg_pool/Pad\"](%/layer3/layer3.2/relu_1/Relu_output_0, %/avg_pool/Constant_output_0), scope: __main__.ResNet34::/torch.nn.modules.pooling.AvgPool2d::avg_pool # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/pooling.py:628:0\n",
" %/avg_pool/AveragePool_output_0 : Float(*, 512, 1, 1, strides=[512, 1, 1, 1], requires_grad=1, device=cpu) = onnx::AveragePool[ceil_mode=0, kernel_shape=[7, 7], pads=[0, 0, 0, 0], strides=[7, 7], onnx_name=\"/avg_pool/AveragePool\"](%/avg_pool/Pad_output_0), scope: __main__.ResNet34::/torch.nn.modules.pooling.AvgPool2d::avg_pool # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/pooling.py:628:0\n",
" %/Shape_output_0 : Long(4, strides=[1], device=cpu) = onnx::Shape[onnx_name=\"/Shape\"](%/avg_pool/AveragePool_output_0), scope: __main__.ResNet34:: # /tmp/ipykernel_33697/1263192908.py:62:0\n",
" %/Constant_output_0 : Long(device=cpu) = onnx::Constant[value={0}, onnx_name=\"/Constant\"](), scope: __main__.ResNet34:: # /tmp/ipykernel_33697/1263192908.py:62:0\n",
" %/Gather_output_0 : Long(device=cpu) = onnx::Gather[axis=0, onnx_name=\"/Gather\"](%/Shape_output_0, %/Constant_output_0), scope: __main__.ResNet34:: # /tmp/ipykernel_33697/1263192908.py:62:0\n",
" %onnx::Unsqueeze_384 : Long(1, strides=[1], device=cpu) = onnx::Constant[value={0}]()\n",
" %/Unsqueeze_output_0 : Long(1, strides=[1], device=cpu) = onnx::Unsqueeze[onnx_name=\"/Unsqueeze\"](%/Gather_output_0, %onnx::Unsqueeze_384), scope: __main__.ResNet34::\n",
" %/Constant_1_output_0 : Long(1, strides=[1], requires_grad=0, device=cpu) = onnx::Constant[value={-1}, onnx_name=\"/Constant_1\"](), scope: __main__.ResNet34::\n",
" %/Concat_output_0 : Long(2, strides=[1], device=cpu) = onnx::Concat[axis=0, onnx_name=\"/Concat\"](%/Unsqueeze_output_0, %/Constant_1_output_0), scope: __main__.ResNet34:: # /tmp/ipykernel_33697/1263192908.py:62:0\n",
" %/Reshape_output_0 : Float(*, 512, strides=[512, 1], requires_grad=1, device=cpu) = onnx::Reshape[allowzero=0, onnx_name=\"/Reshape\"](%/avg_pool/AveragePool_output_0, %/Concat_output_0), scope: __main__.ResNet34:: # /tmp/ipykernel_33697/1263192908.py:62:0\n",
" %/fc/Gemm_output_0 : Float(*, 450, strides=[450, 1], requires_grad=1, device=cpu) = onnx::Gemm[alpha=1., beta=1., transB=1, onnx_name=\"/fc/Gemm\"](%/Reshape_output_0, %fc.weight, %fc.bias), scope: __main__.ResNet34::/torch.nn.modules.linear.Linear::fc # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/modules/linear.py:114:0\n",
" %output : Float(*, 450, strides=[450, 1], requires_grad=1, device=cpu) = onnx::LogSoftmax[axis=1, onnx_name=\"/LogSoftmax\"](%/fc/Gemm_output_0), scope: __main__.ResNet34:: # /home/gautham/.local/lib/python3.10/site-packages/torch/nn/functional.py:1927:0\n",
" return (%output)\n",
"\n"
]
}
],
"source": [
"onnx_path = '../model/model.onnx'\n",
"\n",
"torch.onnx.export(model,\n",
" dummy_input,\n",
" onnx_path,\n",
" verbose=True,\n",
" input_names = ['input'], # the model's input names\n",
" output_names = ['output'], # the model's output names\n",
" dynamic_axes={'input' : {0 : 'batch_size'}, # variable length axes\n",
" 'output' : {0 : 'batch_size'}})"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Verifying the ONNX model"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [],
"source": [
"import onnx\n",
"\n",
"onnx_model = onnx.load(onnx_path)\n",
"onnx.checker.check_model(onnx_model)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Comparing ONNX Runtime and PyTorch results"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Exported model has been tested with ONNXRuntime, and the result looks good!\n"
]
}
],
"source": [
"import onnxruntime\n",
"import numpy as np\n",
"\n",
"ort_session = onnxruntime.InferenceSession(onnx_path)\n",
"\n",
"def to_numpy(tensor):\n",
" return tensor.detach().cpu().numpy() if tensor.requires_grad else tensor.cpu().numpy()\n",
"\n",
"ort_inputs = {ort_session.get_inputs()[0].name: to_numpy(dummy_input)}\n",
"ort_outs = ort_session.run(None, ort_inputs)\n",
"\n",
"np.testing.assert_allclose(to_numpy(torch_out), ort_outs[0], rtol=1e-03, atol=1e-05)\n",
"\n",
"print(\"Exported model has been tested with ONNXRuntime, and the result looks good!\")\n"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3.10.6 64-bit",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.6"
},
"orig_nbformat": 4,
"vscode": {
"interpreter": {
"hash": "916dbcbb3f70747c44a77c7bcd40155683ae19c65e1c03b4aa3499c5328201f1"
}
}
},
"nbformat": 4,
"nbformat_minor": 2
}
|