尝试加载已保存的模型时出现以下错误。
KeyError: 'unexpected key "module.encoder.embedding.weight" in state_dict'
这是我用来加载保存模型的函数。

def load_model_states(model, tag):
    """Load a previously saved model states."""
    filename = os.path.join(args.save_path, tag)
    with open(filename, 'rb') as f:
        model.load_state_dict(torch.load(f))

该模型是一个序列到序列网络,其 init 函数(构造函数)如下所示。
def __init__(self, dictionary, embedding_index, max_sent_length, args):
    """"Constructor of the class."""
    super(Sequence2Sequence, self).__init__()
    self.dictionary = dictionary
    self.embedding_index = embedding_index
    self.config = args
    self.encoder = Encoder(len(self.dictionary), self.config)
    self.decoder = AttentionDecoder(len(self.dictionary), max_sent_length, self.config)
    self.criterion = nn.NLLLoss()  # Negative log-likelihood loss

    # Initializing the weight parameters for the embedding layer in the encoder.
    self.encoder.init_embedding_weights(self.dictionary, self.embedding_index, self.config.emsize)

当我打印模型(序列到序列网络)时,我得到以下信息。
Sequence2Sequence (
  (encoder): Encoder (
    (drop): Dropout (p = 0.25)
    (embedding): Embedding(43723, 300)
    (rnn): LSTM(300, 300, batch_first=True, dropout=0.25)
  )
  (decoder): AttentionDecoder (
    (embedding): Embedding(43723, 300)
    (attn): Linear (600 -> 12)
    (attn_combine): Linear (600 -> 300)
    (drop): Dropout (p = 0.25)
    (out): Linear (300 -> 43723)
    (rnn): LSTM(300, 300, batch_first=True, dropout=0.25)
  )
  (criterion): NLLLoss (
  )
)

所以,module.encoder.embedding 是一个嵌入层,module.encoder.embedding.weight 表示相关的权重矩阵。那么,为什么它说 - unexpected key "module.encoder.embedding.weight" in state_dict

最佳答案

我解决了这个问题。实际上我是使用 nn.DataParallel 保存模型,它将模型存储在模块中,然后我试图在没有 DataParallel 的情况下加载它。因此,要么我需要在我的网络中临时添加一个 nn.DataParallel 以用于加载目的,要么我可以加载权重文件,创建一个没有模块前缀的新有序字典,然后将其加载回来。

第二种解决方法如下所示。

# original saved file with DataParallel
state_dict = torch.load('myfile.pth.tar')
# create new OrderedDict that does not contain `module.`
from collections import OrderedDict
new_state_dict = OrderedDict()
for k, v in state_dict.items():
    name = k[7:] # remove `module.`
    new_state_dict[name] = v
# load params
model.load_state_dict(new_state_dict)

引用:https://discuss.pytorch.org/t/solved-keyerror-unexpected-key-module-encoder-embedding-weight-in-state-dict/1686

关于pytorch - KeyError : 'unexpected key "module. encoder.embedding.weight"in state_dict',我们在Stack Overflow上找到一个类似的问题:https://stackoverflow.com/questions/44230907/

10-12 05:19