当我改为6分类时报错

[xie199389]

Traceback (most recent call last): File "D:/python project/study/CNN/ssd-pytorch-flower/train.py", line 37, in model.load_state_dict(pretrained_dict) File "D:\工作\python3.6\lib\site-packages\torch\nn\modules\module.py", line 830, in load_state_dict self.class.name, "\n\t".join(error_msgs))) RuntimeError: Error(s) in loading state_dict for SSD: size mismatch for conf.0.weight: copying a param with shape torch.Size([84, 512, 3, 3]) from checkpoint, the shape in current model is torch.Size([28, 512, 3, 3]). size mismatch for conf.0.bias: copying a param with shape torch.Size([84]) from checkpoint, the shape in current model is torch.Size([28]). size mismatch for conf.1.weight: copying a param with shape torch.Size([126, 1024, 3, 3]) from checkpoint, the shape in current model is torch.Size([42, 1024, 3, 3]). size mismatch for conf.1.bias: copying a param with shape torch.Size([126]) from checkpoint, the shape in current model is torch.Size([42]). size mismatch for conf.2.weight: copying a param with shape torch.Size([126, 512, 3, 3]) from checkpoint, the shape in current model is torch.Size([42, 512, 3, 3]). size mismatch for conf.2.bias: copying a param with shape torch.Size([126]) from checkpoint, the shape in current model is torch.Size([42]). size mismatch for conf.3.weight: copying a param with shape torch.Size([126, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([42, 256, 3, 3]). size mismatch for conf.3.bias: copying a param with shape torch.Size([126]) from checkpoint, the shape in current model is torch.Size([42]). size mismatch for conf.4.weight: copying a param with shape torch.Size([84, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([28, 256, 3, 3]). size mismatch for conf.4.bias: copying a param with shape torch.Size([84]) from checkpoint, the shape in current model is torch.Size([28]). size mismatch for conf.5.weight: copying a param with shape torch.Size([84, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([28, 256, 3, 3]). size mismatch for conf.5.bias: copying a param with shape torch.Size([84]) from checkpoint, the shape in current model is torch.Size([28]). 当我改为6分类问题时报错如上，已修改config.py,voc_annotation.py,voc_classes.txt中的class类别，但仍然报错，请问是ssd_weights.pth的问题吗，应如何解决？

[bubbliiiing]

已经修改，重新复制一下train文件的内容

[xie199389]

已经修改，重新复制一下train文件的内容

老哥，我看你把model.load_state_dict(pretrained_dict)改为model.load_state_dict(model_dict)，但似乎还是报这个错，除非把config.py的num_classes改回21就不报错了，另外，大神能留个联系方式吗

[bubbliiiing]

有一句忘了复制了 pretrained_dict = {k: v for k, v in pretrained_dict.items() if np.shape(model_dict[k]) == np.shape(v)}

[xie199389]

有一句忘了复制了 pretrained_dict = {k: v for k, v in pretrained_dict.items() if np.shape(model_dict[k]) == np.shape(v)}

谢谢，现在train.py的训练没有问题，但是predict.py预测仍然报错如下： Using TensorFlow backend. 2020-03-26 14:51:12.004041: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_100.dll'; dlerror: cudart64_100.dll not found 2020-03-26 14:51:12.004170: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine. Traceback (most recent call last): File "D:/python project/study/CNN/ssd-pytorch-flower/predict.py", line 5, in ssd = SSD() File "D:\python project\study\CNN\ssd-pytorch-flower\ssd.py", line 37, in init self.generate() File "D:\python project\study\CNN\ssd-pytorch-flower\ssd.py", line 62, in generate model.load_state_dict(torch.load(self.model_path,map_location='cpu')) File "D:\工作\python3.6\lib\site-packages\torch\nn\modules\module.py", line 830, in load_state_dict self.class.name, "\n\t".join(error_msgs))) RuntimeError: Error(s) in loading state_dict for SSD: size mismatch for conf.0.weight: copying a param with shape torch.Size([84, 512, 3, 3]) from checkpoint, the shape in current model is torch.Size([28, 512, 3, 3]). size mismatch for conf.0.bias: copying a param with shape torch.Size([84]) from checkpoint, the shape in current model is torch.Size([28]). size mismatch for conf.1.weight: copying a param with shape torch.Size([126, 1024, 3, 3]) from checkpoint, the shape in current model is torch.Size([42, 1024, 3, 3]). size mismatch for conf.1.bias: copying a param with shape torch.Size([126]) from checkpoint, the shape in current model is torch.Size([42]). size mismatch for conf.2.weight: copying a param with shape torch.Size([126, 512, 3, 3]) from checkpoint, the shape in current model is torch.Size([42, 512, 3, 3]). size mismatch for conf.2.bias: copying a param with shape torch.Size([126]) from checkpoint, the shape in current model is torch.Size([42]). size mismatch for conf.3.weight: copying a param with shape torch.Size([126, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([42, 256, 3, 3]). size mismatch for conf.3.bias: copying a param with shape torch.Size([126]) from checkpoint, the shape in current model is torch.Size([42]). size mismatch for conf.4.weight: copying a param with shape torch.Size([84, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([28, 256, 3, 3]). size mismatch for conf.4.bias: copying a param with shape torch.Size([84]) from checkpoint, the shape in current model is torch.Size([28]). size mismatch for conf.5.weight: copying a param with shape torch.Size([84, 256, 3, 3]) from checkpoint, the shape in current model is torch.Size([28, 256, 3, 3]). size mismatch for conf.5.bias: copying a param with shape torch.Size([84]) from checkpoint, the shape in current model is torch.Size([28]).

[bubbliiiing]

你这里看起来还是模型不匹配诶，你的训练的模型是不是num_classes忘了改，因为你这个是21类的shape

[zkm98]

我感觉如果想去训练自己的模型需要把 model_dict.update(pretrained_dict) 这一句注释掉

[zkm98]

我注释过之后我报错这个情况 Epoch:1/50 iter:0/3 || Loc_Loss: 3.2771 || Conf_Loss: 7.9131 || Traceback (most recent call last): File "马赛克/code/train.py", line 82, in loss_l, loss_c = criterion(out, targets) File "马赛克\torch\nn\modules\module.py", line 532, in call result = self.forward(*input, **kwargs) File "马赛克code\nets\ssd_training.py", line 80, in forward loss_c = log_sum_exp(batch_conf) - batch_conf.gather(1, conf_t.view(-1, 1)) RuntimeError: Invalid index in gather at C:\w\1\s\tmp_conda_3.7_100118\conda\conda-bld\pytorch_1579082551706\work\aten\src\TH/generic/THTensorEvenMoreMath.cpp:657

[bubbliiiing]

我觉得是你的设置出了问题，你自己看错误也知道，你把一个6类的模型导入到20类里面

[zkm98]

我这里的conf函数确实是更改了，确实是我设置的的num_Classes,而且我也打印了看看

[zkm98]

我找到原因了就是在你去计算conf_t的时候没有很好的计算好

[bubbliiiing]

好吧

[zkm98]

有些奇怪你這裏match更新conf_t的時候出現一些數據大於num_classes的情況

[zkm98]

補充一下，上面有一個問題是，應該在類裏面設置背景。不然出現一些問題