训练时网络参数报错怎么解决？

[Eileen000]

训练时网络参数报错怎么解决？

[Eileen000]

这个报错呢？
Traceback (most recent call last):
File "/drive/MyDrive/DANet-0.5.0/danet/train.py", line 204, in
trainer.training(epoch)
File "/drive/MyDrive/DANet-0.5.0/danet/train.py", line 128, in training
outputs = self.model(image)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/parallel/data_parallel.py", line 141, in forward
return self.module(*inputs[0], **kwargs[0])
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/drive/MyDrive/DANet-0.5.0/encoding/models/danet.py", line 45, in forward
_, _, c3, c4 = self.base_forward(x)
File "/drive/MyDrive/DANet-0.5.0/encoding/models/base.py", line 61, in base_forward
c1 = self.pretrained.layer1(x)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/container.py", line 92, in forward
input = module(input)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/drive/MyDrive/DANet-0.5.0/encoding/dilated/resnet.py", line 96, in forward
out = self.conv1(x)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/module.py", line 489, in call
result = self.forward(*input, **kwargs)
File "/usr/local/lib/python3.7/dist-packages/torch/nn/modules/conv.py", line 320, in forward
self.padding, self.dilation, self.groups)
RuntimeError: Given groups=1, weight of size [64, 128, 1, 1], expected input[1, 64, 192, 192] to have 128 channels, but got 64 channels instead

[JIA-HONG-CHU]

在danet.py 更改input 可以使用 transpose 和 view套件

[wlj567]

在 danet.py 更改输入可以使用转置和视图套件
您好，能具体点吗，谢谢了

[JIA-HONG-CHU]

https://zhuanlan.zhihu.com/p/342675997 。這篇有講蠻淺顯易懂的

[wlj567]

https://zhuanlan.zhihu.com/p/342675997 。這篇有講蠻淺顯易懂的

麻烦了，谢谢

[wlj567]

@Eileen000 您好，请问你能运行完整代码吗，方便加个联系方式吗，谢谢了

[Eileen000]

能运行 需要改一些地方 还要指定pytorch版本 以及你的显卡配置不能太低 我有在本地电脑和colab上运行过 需要代码可以 但有偿 ------ 原始邮件 ------ ***@***.***>; 发送时间:2021年9月29日(星期三) 上午9:16 ***@***.***>; ***@***.******@***.***>; 主题:Re: [junfu1115/DANet] 训练时网络参数报错怎么解决？ (#133) @Eileen000 您好，请问你能运行完整代码吗，方便加个联系方式吗，谢谢了 — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.

[JIA-HONG-CHU]

我有用 encnet danet dranet 用在 mmsegmentation 套件上並且套用swin transformer backbone而不是resnet101，你可以參考 https://github.com/JIA-HONG-CHU/Swin-Transformer-add-EncNet-DaNet-DraNet-for-semantic-segmentation-on-Statelite-Dataset

[wlj567]

十分感谢你们的回复，作为新人，代码上的问题颇多，我再好好琢磨琢磨吧。打扰了

[wlj567]

十分感谢你们的回复，作为新人，代码上的问题颇多，我再好好琢磨琢磨吧。打扰了

[wlj567]

您们好，我想问问训练DANet，怎么用cityscapes数据集进行训练？希望能有空回复一下，先谢谢了。