JackyWang-001
commented
1 year ago 新手学习中光是搭环境就用了好久时间，然后好像运行成功了训练可是又没找到如果训练成功了生成的模型在哪里
(fast38) PS C:\Users\Administrator\Desktop\code\python\FastestDet-main> python train.py --yaml weights/train/20230413-2-187/coco.yaml Load yaml sucess... <utils.tool.LoadYaml object at 0x0000026AC27A2D30> Initialize params from:./module/shufflenetv2.pth
Layer (type) Output Shape Param #================================================================
Conv2d-1 [-1, 24, 176, 176] 648 BatchNorm2d-2 [-1, 24, 176, 176] 48 ReLU-3 [-1, 24, 176, 176] 0 MaxPool2d-4 [-1, 24, 88, 88] 0 Conv2d-5 [-1, 24, 44, 44] 216 BatchNorm2d-6 [-1, 24, 44, 44] 48 Conv2d-7 [-1, 24, 44, 44] 576 BatchNorm2d-8 [-1, 24, 44, 44] 48 ReLU-9 [-1, 24, 44, 44] 0 Conv2d-10 [-1, 24, 88, 88] 576 BatchNorm2d-11 [-1, 24, 88, 88] 48 ReLU-12 [-1, 24, 88, 88] 0 Conv2d-13 [-1, 24, 44, 44] 216 BatchNorm2d-14 [-1, 24, 44, 44] 48 Conv2d-15 [-1, 24, 44, 44] 576 BatchNorm2d-16 [-1, 24, 44, 44] 48 ReLU-17 [-1, 24, 44, 44] 0 ShuffleV2Block-18 [-1, 48, 44, 44] 0 Conv2d-19 [-1, 24, 44, 44] 576 BatchNorm2d-20 [-1, 24, 44, 44] 48 ReLU-21 [-1, 24, 44, 44] 0 Conv2d-22 [-1, 24, 44, 44] 216 BatchNorm2d-23 [-1, 24, 44, 44] 48 Conv2d-24 [-1, 24, 44, 44] 576 BatchNorm2d-25 [-1, 24, 44, 44] 48 ReLU-26 [-1, 24, 44, 44] 0 ShuffleV2Block-27 [-1, 48, 44, 44] 0 Conv2d-28 [-1, 24, 44, 44] 576 BatchNorm2d-29 [-1, 24, 44, 44] 48 ReLU-30 [-1, 24, 44, 44] 0 Conv2d-31 [-1, 24, 44, 44] 216 BatchNorm2d-32 [-1, 24, 44, 44] 48 Conv2d-33 [-1, 24, 44, 44] 576 BatchNorm2d-34 [-1, 24, 44, 44] 48 ReLU-35 [-1, 24, 44, 44] 0 ShuffleV2Block-36 [-1, 48, 44, 44] 0 Conv2d-37 [-1, 24, 44, 44] 576 BatchNorm2d-38 [-1, 24, 44, 44] 48 ReLU-39 [-1, 24, 44, 44] 0 Conv2d-40 [-1, 24, 44, 44] 216 BatchNorm2d-41 [-1, 24, 44, 44] 48 Conv2d-42 [-1, 24, 44, 44] 576 BatchNorm2d-43 [-1, 24, 44, 44] 48 ReLU-44 [-1, 24, 44, 44] 0 ShuffleV2Block-45 [-1, 48, 44, 44] 0 Conv2d-46 [-1, 48, 22, 22] 432 BatchNorm2d-47 [-1, 48, 22, 22] 96 Conv2d-48 [-1, 48, 22, 22] 2,304 BatchNorm2d-49 [-1, 48, 22, 22] 96 ReLU-50 [-1, 48, 22, 22] 0 Conv2d-51 [-1, 48, 44, 44] 2,304 BatchNorm2d-52 [-1, 48, 44, 44] 96 ReLU-53 [-1, 48, 44, 44] 0 Conv2d-54 [-1, 48, 22, 22] 432 BatchNorm2d-55 [-1, 48, 22, 22] 96 Conv2d-56 [-1, 48, 22, 22] 2,304 BatchNorm2d-57 [-1, 48, 22, 22] 96 ReLU-58 [-1, 48, 22, 22] 0 ShuffleV2Block-59 [-1, 96, 22, 22] 0 Conv2d-60 [-1, 48, 22, 22] 2,304 BatchNorm2d-61 [-1, 48, 22, 22] 96 ReLU-62 [-1, 48, 22, 22] 0 Conv2d-63 [-1, 48, 22, 22] 432 BatchNorm2d-64 [-1, 48, 22, 22] 96 Conv2d-65 [-1, 48, 22, 22] 2,304 BatchNorm2d-66 [-1, 48, 22, 22] 96 ReLU-67 [-1, 48, 22, 22] 0 ShuffleV2Block-68 [-1, 96, 22, 22] 0 Conv2d-69 [-1, 48, 22, 22] 2,304 BatchNorm2d-70 [-1, 48, 22, 22] 96 ReLU-71 [-1, 48, 22, 22] 0 Conv2d-72 [-1, 48, 22, 22] 432 BatchNorm2d-73 [-1, 48, 22, 22] 96 Conv2d-74 [-1, 48, 22, 22] 2,304 BatchNorm2d-75 [-1, 48, 22, 22] 96 ReLU-76 [-1, 48, 22, 22] 0 ShuffleV2Block-77 [-1, 96, 22, 22] 0 Conv2d-78 [-1, 48, 22, 22] 2,304 BatchNorm2d-79 [-1, 48, 22, 22] 96 ReLU-80 [-1, 48, 22, 22] 0 Conv2d-81 [-1, 48, 22, 22] 432 BatchNorm2d-82 [-1, 48, 22, 22] 96 Conv2d-83 [-1, 48, 22, 22] 2,304 BatchNorm2d-84 [-1, 48, 22, 22] 96 ReLU-85 [-1, 48, 22, 22] 0 ShuffleV2Block-86 [-1, 96, 22, 22] 0 Conv2d-87 [-1, 48, 22, 22] 2,304 BatchNorm2d-88 [-1, 48, 22, 22] 96 ReLU-89 [-1, 48, 22, 22] 0 Conv2d-90 [-1, 48, 22, 22] 432 BatchNorm2d-91 [-1, 48, 22, 22] 96 Conv2d-92 [-1, 48, 22, 22] 2,304 BatchNorm2d-93 [-1, 48, 22, 22] 96 ReLU-94 [-1, 48, 22, 22] 0 ShuffleV2Block-95 [-1, 96, 22, 22] 0 Conv2d-96 [-1, 48, 22, 22] 2,304 BatchNorm2d-97 [-1, 48, 22, 22] 96 ReLU-98 [-1, 48, 22, 22] 0 Conv2d-99 [-1, 48, 22, 22] 432 BatchNorm2d-100 [-1, 48, 22, 22] 96 Conv2d-101 [-1, 48, 22, 22] 2,304 BatchNorm2d-102 [-1, 48, 22, 22] 96 ReLU-103 [-1, 48, 22, 22] 0 ShuffleV2Block-104 [-1, 96, 22, 22] 0 Conv2d-105 [-1, 48, 22, 22] 2,304 BatchNorm2d-106 [-1, 48, 22, 22] 96 ReLU-107 [-1, 48, 22, 22] 0 Conv2d-108 [-1, 48, 22, 22] 432 BatchNorm2d-109 [-1, 48, 22, 22] 96 Conv2d-110 [-1, 48, 22, 22] 2,304 BatchNorm2d-111 [-1, 48, 22, 22] 96 ReLU-112 [-1, 48, 22, 22] 0 ShuffleV2Block-113 [-1, 96, 22, 22] 0 Conv2d-114 [-1, 48, 22, 22] 2,304 BatchNorm2d-115 [-1, 48, 22, 22] 96 ReLU-116 [-1, 48, 22, 22] 0 Conv2d-117 [-1, 48, 22, 22] 432 BatchNorm2d-118 [-1, 48, 22, 22] 96 Conv2d-119 [-1, 48, 22, 22] 2,304 BatchNorm2d-120 [-1, 48, 22, 22] 96 ReLU-121 [-1, 48, 22, 22] 0 ShuffleV2Block-122 [-1, 96, 22, 22] 0 Conv2d-123 [-1, 96, 11, 11] 864 BatchNorm2d-124 [-1, 96, 11, 11] 192 Conv2d-125 [-1, 96, 11, 11] 9,216 BatchNorm2d-126 [-1, 96, 11, 11] 192 ReLU-127 [-1, 96, 11, 11] 0 Conv2d-128 [-1, 96, 22, 22] 9,216 BatchNorm2d-129 [-1, 96, 22, 22] 192 ReLU-130 [-1, 96, 22, 22] 0 Conv2d-131 [-1, 96, 11, 11] 864 BatchNorm2d-132 [-1, 96, 11, 11] 192 Conv2d-133 [-1, 96, 11, 11] 9,216 BatchNorm2d-134 [-1, 96, 11, 11] 192 ReLU-135 [-1, 96, 11, 11] 0 ShuffleV2Block-136 [-1, 192, 11, 11] 0 Conv2d-137 [-1, 96, 11, 11] 9,216 BatchNorm2d-138 [-1, 96, 11, 11] 192 ReLU-139 [-1, 96, 11, 11] 0 Conv2d-140 [-1, 96, 11, 11] 864 BatchNorm2d-141 [-1, 96, 11, 11] 192 Conv2d-142 [-1, 96, 11, 11] 9,216 BatchNorm2d-143 [-1, 96, 11, 11] 192 ReLU-144 [-1, 96, 11, 11] 0 ShuffleV2Block-145 [-1, 192, 11, 11] 0 Conv2d-146 [-1, 96, 11, 11] 9,216 BatchNorm2d-147 [-1, 96, 11, 11] 192 ReLU-148 [-1, 96, 11, 11] 0 Conv2d-149 [-1, 96, 11, 11] 864 BatchNorm2d-150 [-1, 96, 11, 11] 192 Conv2d-151 [-1, 96, 11, 11] 9,216 BatchNorm2d-152 [-1, 96, 11, 11] 192 ReLU-153 [-1, 96, 11, 11] 0 ShuffleV2Block-154 [-1, 192, 11, 11] 0 Conv2d-155 [-1, 96, 11, 11] 9,216 BatchNorm2d-156 [-1, 96, 11, 11] 192 ReLU-157 [-1, 96, 11, 11] 0 Conv2d-158 [-1, 96, 11, 11] 864 BatchNorm2d-159 [-1, 96, 11, 11] 192 Conv2d-160 [-1, 96, 11, 11] 9,216 BatchNorm2d-161 [-1, 96, 11, 11] 192 ReLU-162 [-1, 96, 11, 11] 0 ShuffleV2Block-163 [-1, 192, 11, 11] 0 ShuffleNetV2-164 [[-1, 48, 44, 44], [-1, 96, 22, 22], [-1, 192, 11, 11]] 0 Upsample-165 [-1, 192, 22, 22] 0 AvgPool2d-166 [-1, 48, 22, 22] 0 Conv2d-167 [-1, 96, 22, 22] 32,256 BatchNorm2d-168 [-1, 96, 22, 22] 192 ReLU-169 [-1, 96, 22, 22] 0 Conv1x1-170 [-1, 96, 22, 22] 0 Conv2d-171 [-1, 96, 22, 22] 2,400 BatchNorm2d-172 [-1, 96, 22, 22] 192 ReLU-173 [-1, 96, 22, 22] 0 Conv2d-174 [-1, 96, 22, 22] 2,400 BatchNorm2d-175 [-1, 96, 22, 22] 192 ReLU-176 [-1, 96, 22, 22] 0 Conv2d-177 [-1, 96, 22, 22] 2,400 BatchNorm2d-178 [-1, 96, 22, 22] 192 ReLU-179 [-1, 96, 22, 22] 0 Conv2d-180 [-1, 96, 22, 22] 2,400 BatchNorm2d-181 [-1, 96, 22, 22] 192 ReLU-182 [-1, 96, 22, 22] 0 Conv2d-183 [-1, 96, 22, 22] 2,400 BatchNorm2d-184 [-1, 96, 22, 22] 192 ReLU-185 [-1, 96, 22, 22] 0 Conv2d-186 [-1, 96, 22, 22] 2,400 BatchNorm2d-187 [-1, 96, 22, 22] 192 ReLU-188 [-1, 96, 22, 22] 0 Conv2d-189 [-1, 96, 22, 22] 27,648 BatchNorm2d-190 [-1, 96, 22, 22] 192 ReLU-191 [-1, 96, 22, 22] 0 SPP-192 [-1, 96, 22, 22] 0 Conv2d-193 [-1, 96, 22, 22] 9,216 BatchNorm2d-194 [-1, 96, 22, 22] 192 ReLU-195 [-1, 96, 22, 22] 0 Conv1x1-196 [-1, 96, 22, 22] 0 Conv2d-197 [-1, 96, 22, 22] 2,400 BatchNorm2d-198 [-1, 96, 22, 22] 192 ReLU-199 [-1, 96, 22, 22] 0 Conv2d-200 [-1, 1, 22, 22] 96 BatchNorm2d-201 [-1, 1, 22, 22] 2 Head-202 [-1, 1, 22, 22] 0 Sigmoid-203 [-1, 1, 22, 22] 0 Conv2d-204 [-1, 96, 22, 22] 2,400 BatchNorm2d-205 [-1, 96, 22, 22] 192 ReLU-206 [-1, 96, 22, 22] 0 Conv2d-207 [-1, 4, 22, 22] 384 BatchNorm2d-208 [-1, 4, 22, 22] 8 Head-209 [-1, 4, 22, 22] 0 Conv2d-210 [-1, 96, 22, 22] 2,400 BatchNorm2d-211 [-1, 96, 22, 22] 192 ReLU-212 [-1, 96, 22, 22] 0 Conv2d-213 [-1, 4, 22, 22] 384 BatchNorm2d-214 [-1, 4, 22, 22] 8 Head-215 [-1, 4, 22, 22] 0 Softmax-216 [-1, 4, 22, 22] 0 DetectHead-217 [-1, 9, 22, 22] 0
Total params: 237,042
Trainable params: 237,042 Non-trainable params: 0
Input size (MB): 1.42
Forward/backward pass size (MB): 14982.11 Params size (MB): 0.90 Estimated Total Size (MB): 14984.44 use SGD optimizer Starting training for 10 epochs... Epoch:0 LR:0.000000 IOU:0.683203 Obj:0.118574 Cls:1.722579 Total:9.085380: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:03<00:00, 3.00s/it] Epoch:1 LR:0.000002 IOU:0.682660 Obj:0.117858 Cls:1.543631 Total:8.890644: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.23s/it] Epoch:2 LR:0.000026 IOU:0.642826 Obj:0.115852 Cls:1.679555 Total:8.675802: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.13s/it] Epoch:3 LR:0.000130 IOU:0.680697 Obj:0.117061 Cls:1.506619 Total:8.825181: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.28s/it] Epoch:4 LR:0.000410 IOU:0.693652 Obj:0.116068 Cls:1.497365 Total:8.903667: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.25s/it] Epoch:5 LR:0.001000 IOU:0.667514 Obj:0.118184 Cls:1.399224 Total:8.630288: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.17s/it] Epoch:6 LR:0.001000 IOU:0.641047 Obj:0.113830 Cls:1.065889 Total:8.015538: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.14s/it] Epoch:7 LR:0.001000 IOU:0.598921 Obj:0.112864 Cls:0.718651 Total:7.315843: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.23s/it] Epoch:8 LR:0.001000 IOU:0.596256 Obj:0.111394 Cls:0.472929 Total:7.025274: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.25s/it] Epoch:9 LR:0.001000 IOU:0.594581 Obj:0.108997 Cls:0.344135 Total:6.844735: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.38s/it] Epoch:10 LR:0.001000 IOU:0.586639 Obj:0.108114 Cls:0.251320 Total:6.674245: 100%|███████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.29s/it] computer mAP... 0%| | 0/1 [00:00<?, ?it/s]D :\python\anaconda\envs\fast38\lib\site-packages\torch\functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at C:\actions-runner_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen\native\TensorShape.cpp:2228.) return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined] 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:02<00:00, 2.46s/it] creating index... index created! creating index... index created! Running per image evaluation... Evaluate annotation type bbox DONE (t=0.03s). Accumulating evaluation results... DONE (t=0.01s). Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000
训练应该没问题，在稳步收敛。
模型验证及保存
if epoch % 10 == 0 and epoch > 0:
模型评估
self.model.eval()
print("computer mAP...")
mAP05 = self.evaluation.compute_map(self.val_dataloader, self.model)
torch.save(self.model.state_dict(), "checkpoint/weight_AP05:%f_%d-epoch.pth"%(mAP05, epoch))这部分代码是保存权重的，设置没10轮保存一次，保存在checkpoint文件夹下。当然你可以根据自己的需求，对这些代码进行修改。
summm
touchinglie
新手学习中光是搭环境就用了好久时间，然后好像运行成功了训练可是又没找到如果训练成功了生成的模型在哪里 (fast38) PS C:\Users\Administrator\Desktop\code\python\FastestDet-main> python train.py --yaml weights/train/20230413-2-187/coco.yaml Load yaml sucess... <utils.tool.LoadYaml object at 0x0000026AC27A2D30> Initialize params from:./module/shufflenetv2.pth
================================================================ Conv2d-1 [-1, 24, 176, 176] 648 BatchNorm2d-2 [-1, 24, 176, 176] 48 ReLU-3 [-1, 24, 176, 176] 0 MaxPool2d-4 [-1, 24, 88, 88] 0 Conv2d-5 [-1, 24, 44, 44] 216 BatchNorm2d-6 [-1, 24, 44, 44] 48 Conv2d-7 [-1, 24, 44, 44] 576 BatchNorm2d-8 [-1, 24, 44, 44] 48 ReLU-9 [-1, 24, 44, 44] 0 Conv2d-10 [-1, 24, 88, 88] 576 BatchNorm2d-11 [-1, 24, 88, 88] 48 ReLU-12 [-1, 24, 88, 88] 0 Conv2d-13 [-1, 24, 44, 44] 216 BatchNorm2d-14 [-1, 24, 44, 44] 48 Conv2d-15 [-1, 24, 44, 44] 576 BatchNorm2d-16 [-1, 24, 44, 44] 48 ReLU-17 [-1, 24, 44, 44] 0 ShuffleV2Block-18 [-1, 48, 44, 44] 0 Conv2d-19 [-1, 24, 44, 44] 576 BatchNorm2d-20 [-1, 24, 44, 44] 48 ReLU-21 [-1, 24, 44, 44] 0 Conv2d-22 [-1, 24, 44, 44] 216 BatchNorm2d-23 [-1, 24, 44, 44] 48 Conv2d-24 [-1, 24, 44, 44] 576 BatchNorm2d-25 [-1, 24, 44, 44] 48 ReLU-26 [-1, 24, 44, 44] 0 ShuffleV2Block-27 [-1, 48, 44, 44] 0 Conv2d-28 [-1, 24, 44, 44] 576 BatchNorm2d-29 [-1, 24, 44, 44] 48 ReLU-30 [-1, 24, 44, 44] 0 Conv2d-31 [-1, 24, 44, 44] 216 BatchNorm2d-32 [-1, 24, 44, 44] 48 Conv2d-33 [-1, 24, 44, 44] 576 BatchNorm2d-34 [-1, 24, 44, 44] 48 ReLU-35 [-1, 24, 44, 44] 0 ShuffleV2Block-36 [-1, 48, 44, 44] 0 Conv2d-37 [-1, 24, 44, 44] 576 BatchNorm2d-38 [-1, 24, 44, 44] 48 ReLU-39 [-1, 24, 44, 44] 0 Conv2d-40 [-1, 24, 44, 44] 216 BatchNorm2d-41 [-1, 24, 44, 44] 48 Conv2d-42 [-1, 24, 44, 44] 576 BatchNorm2d-43 [-1, 24, 44, 44] 48 ReLU-44 [-1, 24, 44, 44] 0 ShuffleV2Block-45 [-1, 48, 44, 44] 0 Conv2d-46 [-1, 48, 22, 22] 432 BatchNorm2d-47 [-1, 48, 22, 22] 96 Conv2d-48 [-1, 48, 22, 22] 2,304 BatchNorm2d-49 [-1, 48, 22, 22] 96 ReLU-50 [-1, 48, 22, 22] 0 Conv2d-51 [-1, 48, 44, 44] 2,304 BatchNorm2d-52 [-1, 48, 44, 44] 96 ReLU-53 [-1, 48, 44, 44] 0 Conv2d-54 [-1, 48, 22, 22] 432 BatchNorm2d-55 [-1, 48, 22, 22] 96 Conv2d-56 [-1, 48, 22, 22] 2,304 BatchNorm2d-57 [-1, 48, 22, 22] 96 ReLU-58 [-1, 48, 22, 22] 0 ShuffleV2Block-59 [-1, 96, 22, 22] 0 Conv2d-60 [-1, 48, 22, 22] 2,304 BatchNorm2d-61 [-1, 48, 22, 22] 96 ReLU-62 [-1, 48, 22, 22] 0 Conv2d-63 [-1, 48, 22, 22] 432 BatchNorm2d-64 [-1, 48, 22, 22] 96 Conv2d-65 [-1, 48, 22, 22] 2,304 BatchNorm2d-66 [-1, 48, 22, 22] 96 ReLU-67 [-1, 48, 22, 22] 0 ShuffleV2Block-68 [-1, 96, 22, 22] 0 Conv2d-69 [-1, 48, 22, 22] 2,304 BatchNorm2d-70 [-1, 48, 22, 22] 96 ReLU-71 [-1, 48, 22, 22] 0 Conv2d-72 [-1, 48, 22, 22] 432 BatchNorm2d-73 [-1, 48, 22, 22] 96 Conv2d-74 [-1, 48, 22, 22] 2,304 BatchNorm2d-75 [-1, 48, 22, 22] 96 ReLU-76 [-1, 48, 22, 22] 0 ShuffleV2Block-77 [-1, 96, 22, 22] 0 Conv2d-78 [-1, 48, 22, 22] 2,304 BatchNorm2d-79 [-1, 48, 22, 22] 96 ReLU-80 [-1, 48, 22, 22] 0 Conv2d-81 [-1, 48, 22, 22] 432 BatchNorm2d-82 [-1, 48, 22, 22] 96 Conv2d-83 [-1, 48, 22, 22] 2,304 BatchNorm2d-84 [-1, 48, 22, 22] 96 ReLU-85 [-1, 48, 22, 22] 0 ShuffleV2Block-86 [-1, 96, 22, 22] 0 Conv2d-87 [-1, 48, 22, 22] 2,304 BatchNorm2d-88 [-1, 48, 22, 22] 96 ReLU-89 [-1, 48, 22, 22] 0 Conv2d-90 [-1, 48, 22, 22] 432 BatchNorm2d-91 [-1, 48, 22, 22] 96 Conv2d-92 [-1, 48, 22, 22] 2,304 BatchNorm2d-93 [-1, 48, 22, 22] 96 ReLU-94 [-1, 48, 22, 22] 0 ShuffleV2Block-95 [-1, 96, 22, 22] 0 Conv2d-96 [-1, 48, 22, 22] 2,304 BatchNorm2d-97 [-1, 48, 22, 22] 96 ReLU-98 [-1, 48, 22, 22] 0 Conv2d-99 [-1, 48, 22, 22] 432 BatchNorm2d-100 [-1, 48, 22, 22] 96 Conv2d-101 [-1, 48, 22, 22] 2,304 BatchNorm2d-102 [-1, 48, 22, 22] 96 ReLU-103 [-1, 48, 22, 22] 0 ShuffleV2Block-104 [-1, 96, 22, 22] 0 Conv2d-105 [-1, 48, 22, 22] 2,304 BatchNorm2d-106 [-1, 48, 22, 22] 96 ReLU-107 [-1, 48, 22, 22] 0 Conv2d-108 [-1, 48, 22, 22] 432 BatchNorm2d-109 [-1, 48, 22, 22] 96 Conv2d-110 [-1, 48, 22, 22] 2,304 BatchNorm2d-111 [-1, 48, 22, 22] 96 ReLU-112 [-1, 48, 22, 22] 0 ShuffleV2Block-113 [-1, 96, 22, 22] 0 Conv2d-114 [-1, 48, 22, 22] 2,304 BatchNorm2d-115 [-1, 48, 22, 22] 96 ReLU-116 [-1, 48, 22, 22] 0 Conv2d-117 [-1, 48, 22, 22] 432 BatchNorm2d-118 [-1, 48, 22, 22] 96 Conv2d-119 [-1, 48, 22, 22] 2,304 BatchNorm2d-120 [-1, 48, 22, 22] 96 ReLU-121 [-1, 48, 22, 22] 0 ShuffleV2Block-122 [-1, 96, 22, 22] 0 Conv2d-123 [-1, 96, 11, 11] 864 BatchNorm2d-124 [-1, 96, 11, 11] 192 Conv2d-125 [-1, 96, 11, 11] 9,216 BatchNorm2d-126 [-1, 96, 11, 11] 192 ReLU-127 [-1, 96, 11, 11] 0 Conv2d-128 [-1, 96, 22, 22] 9,216 BatchNorm2d-129 [-1, 96, 22, 22] 192 ReLU-130 [-1, 96, 22, 22] 0 Conv2d-131 [-1, 96, 11, 11] 864 BatchNorm2d-132 [-1, 96, 11, 11] 192 Conv2d-133 [-1, 96, 11, 11] 9,216 BatchNorm2d-134 [-1, 96, 11, 11] 192 ReLU-135 [-1, 96, 11, 11] 0 ShuffleV2Block-136 [-1, 192, 11, 11] 0 Conv2d-137 [-1, 96, 11, 11] 9,216 BatchNorm2d-138 [-1, 96, 11, 11] 192 ReLU-139 [-1, 96, 11, 11] 0 Conv2d-140 [-1, 96, 11, 11] 864 BatchNorm2d-141 [-1, 96, 11, 11] 192 Conv2d-142 [-1, 96, 11, 11] 9,216 BatchNorm2d-143 [-1, 96, 11, 11] 192 ReLU-144 [-1, 96, 11, 11] 0 ShuffleV2Block-145 [-1, 192, 11, 11] 0 Conv2d-146 [-1, 96, 11, 11] 9,216 BatchNorm2d-147 [-1, 96, 11, 11] 192 ReLU-148 [-1, 96, 11, 11] 0 Conv2d-149 [-1, 96, 11, 11] 864 BatchNorm2d-150 [-1, 96, 11, 11] 192 Conv2d-151 [-1, 96, 11, 11] 9,216 BatchNorm2d-152 [-1, 96, 11, 11] 192 ReLU-153 [-1, 96, 11, 11] 0 ShuffleV2Block-154 [-1, 192, 11, 11] 0 Conv2d-155 [-1, 96, 11, 11] 9,216 BatchNorm2d-156 [-1, 96, 11, 11] 192 ReLU-157 [-1, 96, 11, 11] 0 Conv2d-158 [-1, 96, 11, 11] 864 BatchNorm2d-159 [-1, 96, 11, 11] 192 Conv2d-160 [-1, 96, 11, 11] 9,216 BatchNorm2d-161 [-1, 96, 11, 11] 192 ReLU-162 [-1, 96, 11, 11] 0 ShuffleV2Block-163 [-1, 192, 11, 11] 0 ShuffleNetV2-164 [[-1, 48, 44, 44], [-1, 96, 22, 22], [-1, 192, 11, 11]] 0 Upsample-165 [-1, 192, 22, 22] 0 AvgPool2d-166 [-1, 48, 22, 22] 0 Conv2d-167 [-1, 96, 22, 22] 32,256 BatchNorm2d-168 [-1, 96, 22, 22] 192 ReLU-169 [-1, 96, 22, 22] 0 Conv1x1-170 [-1, 96, 22, 22] 0 Conv2d-171 [-1, 96, 22, 22] 2,400 BatchNorm2d-172 [-1, 96, 22, 22] 192 ReLU-173 [-1, 96, 22, 22] 0 Conv2d-174 [-1, 96, 22, 22] 2,400 BatchNorm2d-175 [-1, 96, 22, 22] 192 ReLU-176 [-1, 96, 22, 22] 0 Conv2d-177 [-1, 96, 22, 22] 2,400 BatchNorm2d-178 [-1, 96, 22, 22] 192 ReLU-179 [-1, 96, 22, 22] 0 Conv2d-180 [-1, 96, 22, 22] 2,400 BatchNorm2d-181 [-1, 96, 22, 22] 192 ReLU-182 [-1, 96, 22, 22] 0 Conv2d-183 [-1, 96, 22, 22] 2,400 BatchNorm2d-184 [-1, 96, 22, 22] 192 ReLU-185 [-1, 96, 22, 22] 0 Conv2d-186 [-1, 96, 22, 22] 2,400 BatchNorm2d-187 [-1, 96, 22, 22] 192 ReLU-188 [-1, 96, 22, 22] 0 Conv2d-189 [-1, 96, 22, 22] 27,648 BatchNorm2d-190 [-1, 96, 22, 22] 192 ReLU-191 [-1, 96, 22, 22] 0 SPP-192 [-1, 96, 22, 22] 0 Conv2d-193 [-1, 96, 22, 22] 9,216 BatchNorm2d-194 [-1, 96, 22, 22] 192 ReLU-195 [-1, 96, 22, 22] 0 Conv1x1-196 [-1, 96, 22, 22] 0 Conv2d-197 [-1, 96, 22, 22] 2,400 BatchNorm2d-198 [-1, 96, 22, 22] 192 ReLU-199 [-1, 96, 22, 22] 0 Conv2d-200 [-1, 1, 22, 22] 96 BatchNorm2d-201 [-1, 1, 22, 22] 2 Head-202 [-1, 1, 22, 22] 0 Sigmoid-203 [-1, 1, 22, 22] 0 Conv2d-204 [-1, 96, 22, 22] 2,400 BatchNorm2d-205 [-1, 96, 22, 22] 192 ReLU-206 [-1, 96, 22, 22] 0 Conv2d-207 [-1, 4, 22, 22] 384 BatchNorm2d-208 [-1, 4, 22, 22] 8 Head-209 [-1, 4, 22, 22] 0 Conv2d-210 [-1, 96, 22, 22] 2,400 BatchNorm2d-211 [-1, 96, 22, 22] 192 ReLU-212 [-1, 96, 22, 22] 0 Conv2d-213 [-1, 4, 22, 22] 384 BatchNorm2d-214 [-1, 4, 22, 22] 8 Head-215 [-1, 4, 22, 22] 0 Softmax-216 [-1, 4, 22, 22] 0 DetectHead-217 [-1, 9, 22, 22] 0
Total params: 237,042 Trainable params: 237,042 Non-trainable params: 0
Input size (MB): 1.42 Forward/backward pass size (MB): 14982.11 Params size (MB): 0.90 Estimated Total Size (MB): 14984.44
use SGD optimizer Starting training for 10 epochs... Epoch:0 LR:0.000000 IOU:0.683203 Obj:0.118574 Cls:1.722579 Total:9.085380: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:03<00:00, 3.00s/it] Epoch:1 LR:0.000002 IOU:0.682660 Obj:0.117858 Cls:1.543631 Total:8.890644: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.23s/it] Epoch:2 LR:0.000026 IOU:0.642826 Obj:0.115852 Cls:1.679555 Total:8.675802: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.13s/it] Epoch:3 LR:0.000130 IOU:0.680697 Obj:0.117061 Cls:1.506619 Total:8.825181: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.28s/it] Epoch:4 LR:0.000410 IOU:0.693652 Obj:0.116068 Cls:1.497365 Total:8.903667: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.25s/it] Epoch:5 LR:0.001000 IOU:0.667514 Obj:0.118184 Cls:1.399224 Total:8.630288: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.17s/it] Epoch:6 LR:0.001000 IOU:0.641047 Obj:0.113830 Cls:1.065889 Total:8.015538: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.14s/it] Epoch:7 LR:0.001000 IOU:0.598921 Obj:0.112864 Cls:0.718651 Total:7.315843: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.23s/it] Epoch:8 LR:0.001000 IOU:0.596256 Obj:0.111394 Cls:0.472929 Total:7.025274: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.25s/it] Epoch:9 LR:0.001000 IOU:0.594581 Obj:0.108997 Cls:0.344135 Total:6.844735: 100%|████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.38s/it] Epoch:10 LR:0.001000 IOU:0.586639 Obj:0.108114 Cls:0.251320 Total:6.674245: 100%|███████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00, 1.29s/it] computer mAP... 0%| | 0/1 [00:00<?, ?it/s]D :\python\anaconda\envs\fast38\lib\site-packages\torch\functional.py:568: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at
C:\actions-runner_work\pytorch\pytorch\builder\windows\pytorch\aten\src\ATen\native\TensorShape.cpp:2228.) return _VF.meshgrid(tensors, *kwargs) # type: ignore[attr-defined] 100%|███████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:02<00:00, 2.46s/it] creating index... index created! creating index... index created! Running per image evaluation... Evaluate annotation type bbox* DONE (t=0.03s). Accumulating evaluation results... DONE (t=0.01s). Average Precision (AP) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50 | area= all | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.75 | area= all | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000 Average Precision (AP) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 1 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets= 10 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= all | maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= small | maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area=medium | maxDets=100 ] = 0.000 Average Recall (AR) @[ IoU=0.50:0.95 | area= large | maxDets=100 ] = 0.000