TypeError: __init__() takes 1 positional argument but 2 were given

[兿然]

class BertForMultiLable(BertPreTrainedModel):
def init(self, config):

    super(BertForMultiLable, self).__init__(config)
    self.bert = BertModel(config)
    self.dropout = nn.Dropout(config.hidden_dropout_prob)
    self.classifier = nn.Linear(config.hidden_size, config.num_labels)
    self.apply(self.init_weights)

when we try to run the bert model by your steps, everything is perfect, however there is always a type error showing that the init_weights() function in the line 11

self.apply(self.init_weights)

of bert_for_multi_label.py has something wrong. we tried to modify it to

self.apply(self._init_weights)

class BertForMultiLable(BertPreTrainedModel):
def init(self, config):

    super(BertForMultiLable, self).__init__(config)
    self.bert = BertModel(config)
    self.dropout = nn.Dropout(config.hidden_dropout_prob)
    self.classifier = nn.Linear(config.hidden_size, config.num_labels)
    self.apply(self._init_weights)

then it looks like okay. I am wondering if you also meet this problem and if the problem is caused by the version of python or the inheritance of parent class？

[Jifa Zhou]

when you use the torch 1.0.0, you may meet this problem
and 1.1.0 is ok

[tyj117]

I have tried 1.1.0, but it doesn't work. Have you solved this problem?

[Aditya Vartak]

@FayeXXX just check your pytorch version. if it is >= 1.2.0, do the following change to your init():
modify self.apply(self.init_weights) to self.init_weights()

    super(BertForMultiLable, self).__init__(config)
    self.bert = BertModel(config)
    self.dropout = nn.Dropout(config.hidden_dropout_prob)
    self.classifier = nn.Linear(config.hidden_size, config.num_labels)
    self.init_weights()

@lonePatient has already explained it in the readme of the file.

note: pytorch_transformers>=1.2.0, modify self.apply(self.init_weights) to self.init_weights() in pybert/model/nn/bert_for_multi_label.py file .

Try this fix