search

ShemoonX / Chinese-image-caption

Image Chinese Description Generation Based on Multi-level Selective Visual Semantic Attributes
12 stars 5 forks source link

异常 #9

Open Dabiaoge111 opened 1 year ago

Dabiaoge111 commented 1 year ago

您好，在运行multi_label_classification.py时出现这个异常，请问是哪里出现了问题呢？

RuntimeError: size mismatch, m1: [16 x 8192], m2: [2048 x 2048] at /pytorch/aten/src/THC/generic/THCTensorMathBlas.cu:266

ShemoonX commented 1 year ago

你好，可能要考虑一下多标签分类时，特征提取网络的维度问题。

Dabiaoge111 commented 1 year ago

感谢您的回复，但是我再次查阅了代码之后，还是没找见得从哪里更改维度，这是完整的报错信息。 Traceback (most recent call last): File "multi_label_classification.py", line 132, in train() File "multi_label_classification.py", line 101, in train labels = model(imgs) File "/home/.conda/envs/cnImage/lib/python3.5/site-packages/torch/nn/modules/module.py", line 489, in call result = self.forward(*input, *kwargs) File "/home/.conda/envs/cnImage/lib/python3.5/site-packages/torchvision/models/resnet.py", line 151, in forward x = self.fc(x) File "/home/.conda/envs/cnImage/lib/python3.5/site-packages/torch/nn/modules/module.py", line 489, in call result = self.forward(input, **kwargs) File "/home/.conda/envs/cnImage/lib/python3.5/site-packages/torch/nn/modules/linear.py", line 67, in forward return F.linear(input, self.weight, self.bias) File "/home/.conda/envs/cnImage/lib/python3.5/site-packages/torch/nn/functional.py", line 1352, in linear ret = torch.addmm(torch.jit._unwrap_optional(bias), input, weight.t()) RuntimeError: size mismatch, m1: [16 x 8192], m2: [2048 x 2048] at /pytorch/aten/src/THC/generic/THCTensorMathBlas.cu:266

ShemoonX commented 1 year ago

你好，由于项目时间过久，具体的异常可能不是很清楚了。但multi_label_classification主要是一个多标签分类器，用于将一张输入图像映射到由图像描述提取出的多个（2048个）具象标签。具体可以参考一下论文，论文里有更详细的描述，说不定可以找到相应的解决方案。谢谢。 论文地址：http://jcip.cipsc.org.cn/CN/Y2021/V35/I4/129

Dabiaoge111 commented 1 year ago

好的，再次感谢您的回复。