Can I place my target variable inside the forward function of the data model class

🤗Transformers
1

Hey guys, I’m working on a multioutput classification task.
Can I place my target variable inside the forward function in the data model class i.e in this case

class BERTModel(nn.Module):
def init(self):
super(BERTModel, self).init()
self.bert = BertModel.from_pretrained(‘bert-base-uncased’, return_dict=True)
self.dropout = nn.Dropout(0.3)
self.classified = nn.Linear(768, 6)

def forward(self, ids, mask, token_type_ids, target):
    _, pooled_output = self.bert(
        input_ids = ids,
        attention_mask = mask,
        token_type_ids = token_type_ids,
        target = target 
    )

    pooled_output = self.dropout(pooled_output)
    outputs = self.classified(pooled_output)
    return outputs