models.py

import logging
from typing import Optional, Tuple
from typing import Union

import torch
from torch.nn import CrossEntropyLoss
from transformers import PreTrainedModel, BertForMaskedLM, BertConfig
from transformers.modeling_outputs import SequenceClassifierOutput


class StanceEncoderModel(PreTrainedModel):

    config_class = BertConfig
    logger = logging.getLogger("StanceEncoderModel")

    def __init__(self, config):
        super().__init__(config)
        task_specific_params = config.task_specific_params
        self.num_labels = task_specific_params.get('num_labels', 3)
        self.mask_token_id = task_specific_params['mask_token_id']
        self.verbalizer_token_ids = task_specific_params['verbalizer_token_ids']
        self.clf_hidden_dim = task_specific_params.get('clf_hidden_dim', 300)
        self.clf_drop_prob = task_specific_params.get('clf_drop_prob', 0.2)
        self.clf_gelu_head = task_specific_params.get('clf_gelu_head', False)
        self.masked_lm = task_specific_params.get('masked_lm', True)
        self.masked_lm_n_tokens = task_specific_params.get('masked_lm_tokens', 1)
        self.masked_lm_verbalizer = task_specific_params.get('masked_lm_verbalizer', False)

        base_model = BertForMaskedLM(config)
        self.base_enc_model = base_model.bert
        self.lm_head = base_model.cls

        hidden_size_multiplier = 1

        if not self.masked_lm_verbalizer:
            if self.clf_gelu_head:
                self.logger.info('using 2 layer gelu classifier head')
                self.classifier = torch.nn.Sequential(
                    torch.nn.Linear(self.config.hidden_size * hidden_size_multiplier, self.clf_hidden_dim),
                    torch.nn.Dropout(self.clf_drop_prob),
                    torch.nn.GELU(),
                    torch.nn.Linear(self.clf_hidden_dim, self.num_labels)
                )
            else:
                raise ValueError('classification type head not specified')

    def forward(
            self,
            input_ids: Optional[torch.Tensor] = None,
            attention_mask: Optional[torch.Tensor] = None,
            token_type_ids: Optional[torch.Tensor] = None,
            position_ids: Optional[torch.Tensor] = None,
            head_mask: Optional[torch.Tensor] = None,
            inputs_embeds: Optional[torch.Tensor] = None,
            labels: Optional[torch.Tensor] = None,
            output_attentions: Optional[bool] = None,
            output_hidden_states: Optional[bool] = None,
            return_dict: Optional[bool] = None,
            sequence_ids: Optional[torch.Tensor] = None,
    ) -> Union[Tuple[torch.Tensor], SequenceClassifierOutput]:

        outputs = self.base_enc_model(
            input_ids=input_ids,
            attention_mask=attention_mask
        )

        masked_token_filter = input_ids == self.mask_token_id
        masked_repr = outputs.last_hidden_state[masked_token_filter].reshape(len(input_ids), -1)

        if self.masked_lm_verbalizer:
            logits = self.lm_head(masked_repr)[:, self.verbalizer_token_ids]
        else:
            logits = self.classifier(masked_repr)

        loss = None
        if labels is not None:
            loss_fct = CrossEntropyLoss()
            loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))

        return SequenceClassifierOutput(
            loss=loss,
            logits=logits,
            hidden_states=outputs.hidden_states,
            attentions=outputs.attentions,
        )