ValueError: el Uso de un tamaño de destino (de la antorcha.Tamaño([2, 1])) que es diferente al tamaño de entrada (de la antorcha.Tamaño([16, 1])) está en desuso

Estoy tratando de construir un modelo para la Quora preguntas par de conjunto de datos en el que la salida es binaria, ya sea 1 o 0, pero me sale este error. Sé que la salida de la forma de mi modelo es diferente de la de entrada de la forma, pero no sé cómo solucionarlo. El tamaño del lote es de 16

    class Bert_model (nn.Module):
      def __init__(self) :
        super(Bert_model,self).__init__()
        self.bert =  BertModel.from_pretrained('bert-base-uncased', return_dict=False)
        self.drop_layer = nn.Dropout(.25)
        self.output = nn.Linear(self.bert.config.hidden_size,1)
    
      def forward(self,input_ids,attention_mask):
        _,o2 = self.bert (input_ids =input_ids , attention_mask = attention_mask )
        o2 = self.drop_layer(o2)
        return self.output(o2)

    model = Bert_model()
    
    loss_fn = nn.BCELoss().to(device)

    def train_epoch(
      model, 
      data_loader, 
      loss_fn, 
      optimizer, 
      device, 
      n_examples
    ):
      model = model.train()
    
      losses = []
      correct_predictions = 0
      
      for d in data_loader:
        input_ids = d["input_ids"].to(device)
        attention_mask = d["attention_mask"].to(device)
        targets = d["target"].to(device)
    
        input_ids = input_ids.view(BATCH_SIZE,-1)
        attention_mask = attention_mask.view(BATCH_SIZE,-1)
    
        outputs = model(
          input_ids=input_ids,
          attention_mask=attention_mask
        )
    
        _, preds = torch.max(outputs, dim=1)
    
        targets = targets.unsqueeze(-1)
        loss = loss_fn(F.softmax(outputs,dim=1), targets)
    
        correct_predictions += torch.sum(preds == targets)
        losses.append(loss.item())
    
        loss.backward()
        nn.utils.clip_grad_norm_(model.parameters(), max_norm=1.0)
        optimizer.step()
        optimizer.zero_grad()
    
      return correct_predictions.double() / n_examples, np.mean(losses)

El error:

/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py in
binary_cross_entropy(input, target, weight, size_average, reduce,
reduction)    2913         weight = weight.expand(new_size)    2914 
-> 2915     return torch._C._nn.binary_cross_entropy(input, target, weight, reduction_enum)    2916     2917  ValueError: Using a target
size (torch.Size([2, 1])) that is different to the input size
(torch.Size([16, 1])) is deprecated