RuntimeError: shape '[32, 15, 845]' is invalid for input of size 540800

[Zyt-wenwne]

Hi, Thanks for your work , but when I run the code , a error happened, Traceback (most recent call last): File "train_meta.py", line 352, in train(epoch) File "train_meta.py", line 197, in train loss,recall_rate,conf_proposal,loss_bbox,loss_conf,loss_cls = region_loss(output, target) File "/home/plc319/anaconda3/envs/ct/lib/python3.6/site-packages/torch/nn/modules/module.py", line 532, in call result = self.forward(*input, *kwargs) File "/home/plc319/桌面/Few-shot-Object-Detection-via-Feature-Reweighting-main/models/layers/region_loss.py", line 296, in forward cls = cls.view(bs, cs, nA nC nH nW).transpose(1, 2).contiguous().view(bs nA nC nH nW, cs) RuntimeError: shape '[32, 15, 845]' is invalid for input of size 540800 can you tell me how to solve the problem

[whattoshow]

I have the same problem while training base model, and my enviroment is cuda10.1 torch0.3.1 torchvision0.5.0, is there anyone could help us?

[ZZY-666]

Hi, Thanks for your work , but when I run the code , a error happened, Traceback (most recent call last): File "train_meta.py", line 352, in train(epoch) File "train_meta.py", line 197, in train loss,recall_rate,conf_proposal,loss_bbox,loss_conf,loss_cls = region_loss(output, target) File "/home/plc319/anaconda3/envs/ct/lib/python3.6/site-packages/torch/nn/modules/module.py", line 532, in call result = self.forward(*input, *kwargs) File "/home/plc319/桌面/Few-shot-Object-Detection-via-Feature-Reweighting-main/models/layers/region_loss.py", line 296, in forward cls = cls.view(bs, cs, nA nC nH nW).transpose(1, 2).contiguous().view(bs nA nC nH nW, cs) RuntimeError: shape '[32, 15, 845]' is invalid for input of size 540800 can you tell me how to solve the problem

您好 我遇到了同样的问题 您解决了吗？有偿

[ZZY-666]

I have the same problem while training base model, and my enviroment is cuda10.1 torch0.3.1 torchvision0.5.0, is there anyone could help us?

您好 我遇到了同样的问题 您解决了吗？有偿

[wangzuo-hx]

@ZZY-666 请问你现在解决了吗

[gusdhg]

I solve it by change "gpu=0" in the metayolo.data

[hwhwantlearn]

I have tried to "change GPUs=1,2,3,4" to "GPU=0" in metayolo.data, but it is not solved. Is there any other solution?

[hwhwantlearn]

@gusdhg @ZZY-666 @whattoshow @zuowanggithub May I know if you have solved the problem? thanks.

[keys-zlc]

应该是base类的内容不对，我改过base类的类别后就对了，改成自己的