【PyTorch】加载模型报错：RuntimeError: Error(s) in loading state_dict for ResNet:Missing key(s) in state_dict

HEX9CF

已于 2024-12-24 08:31:35 修改

阅读量773

点赞数 11

分类专栏： PyTorch 文章标签： pytorch 人工智能 python

于 2024-12-24 08:27:25 首次发布

本文链接：https://blog.youkuaiyun.com/qq_34988204/article/details/144636829

版权

PyTorch 专栏收录该内容

7 篇文章

订阅专栏

【PyTorch】加载模型报错：RuntimeError: Error(s) in loading state_dict for ResNet:Missing key(s) in state_dict

问题描述

  resnet.load_state_dict(t.load(PHOTO_RESNET, map_location=t.device('cpu')))
Traceback (most recent call last):
  File "SketchyDatabase\feature\feature_extract.py", line 66, in <module>
    resnet.load_state_dict(t.load(PHOTO_RESNET, map_location=t.device('cpu')))
  File "D:\ProgramData\Anaconda\envs\pytorch\lib\site-packages\torch\nn\modules\module.py", line 2215, in load_state_dict
    raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for ResNet:
	Missing key(s) in state_dict: "conv1.weight", "bn1.weight", "bn1.bias", "bn1.running_mean", "bn1.running_var", "layer1.0.conv1.weight", 
	...

原因分析

加载的模型状态字典中的键名带有 “module.” 前缀，这是因为模型在训练时使用了 torch.nn.DataParallel。

解决方案

将 strict 设为 False：

resnet.load_state_dict(t.load(PHOTO_RESNET, map_location=t.device('cpu'), weights_only=True), strict=False)