您好，我遇到了train.py中的run_fusion函数报错如下：

Linfeng-Tang / SeAFusion

The code of " Image fusion in the loop of high-level vision tasks: A semantic-aware real-time infrared and visible image fusion network"

MIT License

184 stars 34 forks source link

您好，我遇到了train.py中的run_fusion函数报错如下： #31

Open Wbzb opened 9 months ago

Wbzb commented 9 months ago

RuntimeError: Error(s) in loading state_dict for FusionNet: Missing key(s) in state_dict: "decode4.bn.weight", "decode4.bn.bias", "decode4.bn.running_mean", "decode4.bn.running_var", "decode3.bn.weight", "decode3.bn.bias", "decode3.bn.running_mean", "decode3.bn.running_var", "decode2.bn.weight", "decode2.bn.bias", "decode2.bn.running_mean", "decode2.bn.running_var", "decode1.bn.weight", "decode1.bn.bias", "decode1.bn.running_mean", "decode1.bn.running_var".

Linfeng-Tang commented 9 months ago

看这个错误显示需要加载预训练的权重，但是没有加载上，有可能是load了一个错误的融合权重哈。这个需要再检查一下代码。

HaoyuYao commented 8 months ago

需要加载fusionmodel_final.pth才可以