Unexpected key(s) in state_dict: “module.conv1.weight“, “module.bn1.weight“, “module.bn1.bias“,

由于服务器老是断电所以想加载已经训练好的上一个epoch的模型，但是在加载时遇到了这个问题
这是由于保存模型字典时每一个模块的key都自动加上了‘module’。所以在加载模型参数继续训练时就会与模型对不上号。

RuntimeError: Error(s) in loading state_dict for ResNet:Unexpected key(s) in state_dict: "module.conv1.weight", "module.bn1.weight", "module.bn1.bias", "module.bn1.running_mean", "module.bn1.running_var", "module.conv2.weight", "module.bn2.weight", "module.bn2.bias", "module.bn2.running_mean", "module.bn2.running_var", "module.conv3.weight", "module.bn3.weight", "module.bn3.bias", "module.bn3.running_mean", "module.bn3.running_var", "module.layer1.0.conv1.weight", "module.layer1.0.bn1.weight", "module.layer1.0.bn1.bias", "module.layer1.0.bn1.running_mean", "module.layer1.0.bn1.running_var", "module.layer1.0.conv2.weight", "module.layer1.0.bn2.weight", "module.layer1.0.bn2.bias",

接着查看测试的代码：可以看到在测试中就是把’module.'给去掉，从下标7开始读取key。

    for key, nkey in zip(state_dict_old.keys(), state_dict.keys()):if key != nkey:# remove the 'module.' in the 'key'state_dict[key[7:]] = deepcopy(state_dict_old[key])else:state_dict[key] = deepcopy(state_dict_old[key])

所以在测试的时候直接去掉前面的module.就好了像这样：new_param是从上一个epoch读取出的参数字典

    deeplab.load_state_dict({k.replace('module.',''):v for k,v in new_params.items()})

Unexpected key(s) in state_dict: “module.conv1.weight“, “module.bn1.weight“, “module.bn1.bias“,相关推荐

pytorch加载模型报错Unexpected key(s) in state_dict: module.conv1.weight, module.bn1
文章目录背景报错原因解决背景 Pytorch在加载模型参数的时候,有两种情况可能出现这种问题: 自己写的网络结构,例如: 代码 import models arch = 'resnet50' ...
PyTorch加载模型model.load_state_dict()问题，Unexpected key(s) in state_dict: “module.features..，Expected .
希望将训练好的模型加载到新的网络上.如上面题目所描述的,PyTorch在加载之前保存的模型参数的时候,遇到了问题. Unexpected key(s) in state_dict: "mod ...
Unexpected key(s) in state_dict: module.backbone.bn1.num_batches_tracked
pytorch预测的时候报异常了: {RuntimeError}Error(s) in loading state_dict for DataParallel: Unexpected key( ...
Huggingface填坑：Unexpected key(s) in state_dict: “roberta.pooler.dense.weight“....
在加载已经训练好的模型时,出现问题如下: RuntimeError: Error(s) in loading state_dict for RobertaForSequenceClassificati ...
Unexpected key(s) in state_dict: “dense_block1.denselayer1.norm.1
Unexpected key(s) in state_dict: "dense_block1.denselayer1.norm.1 from torchvision.models impor ...
Missing key(s) in state_dict | unexpected_keys | 模型加载 [torch.load] 报错【解决方案】
文章目录 Missing key(s) in state_dict 报错分析该报错快速直接解决方法额外思路 [Unexpected key(s) in state_dict]- 解决思路 Miss ...
Missing key(s) in state_dict
Missing key(s) in state_dict 解决方法: try:from collections import OrderedDictnew_state_dict = OrderedDi ...
【Python】解决CNN中训练权重参数不匹配size mismatch for fc.weight，size mismatch for fc.bias
目录 1.问题描述 2.问题原因 3.问题解决 3.1思路1--忽视最后一层权重额外说明:假如载入权重不写strict=False, 直接是model.load_state_dict(pre_wei ...
[RN] React Native 错误 Module does not exist in the module map
React Native 错误 Module does not exist in the module map 代码如下: import Login from 'login' import Index ...

Unexpected key(s) in state_dict: “module.conv1.weight“, “module.bn1.weight“, “module.bn1.bias“,

Unexpected key(s) in state_dict: “module.conv1.weight“, “module.bn1.weight“, “module.bn1.bias“,相关推荐

最新文章

热门文章