modeling.py

import torch
import torch.nn as nn
from transformers import AutoModel

class BERTMultiLabel(nn.Module):
    def __init__(self, model_name="microsoft/deberta-v3-base", num_labels=5):
        super().__init__()
        self.bert = AutoModel.from_pretrained(model_name)
        hidden = self.bert.config.hidden_size
        self.dropout = nn.Dropout(0.2)
        self.classifier = nn.Linear(hidden, num_labels)

    def forward(self, input_ids, attention_mask):
        outputs = self.bert(
            input_ids=input_ids,
            attention_mask=attention_mask
        )

        cls = outputs.last_hidden_state[:, 0]  # CLS token
        cls = self.dropout(cls)
        logits = self.classifier(cls)
        return logits