from typing import List, Optional, Tuple, Union

import torch
from torch import nn
from transformers.modeling_outputs import TokenClassifierOutput
from transformers.models.roberta.modeling_roberta import RobertaForTokenClassification
from transformers.models.roberta.modeling_roberta import ROBERTA_INPUTS_DOCSTRING, add_start_docstrings_to_model_forward, add_code_sample_docstrings

from extended_embeddings.extended_embeddings_model import ExtendedEmbeddigsRobertaModel

_CONFIG_FOR_DOC = "RobertaConfig"


class ExtendedEmbeddigsRobertaForTokenClassification(RobertaForTokenClassification):
    def __init__(self, config):
        super().__init__(config)
        self.num_labels = config.num_labels

        self.roberta = ExtendedEmbeddigsRobertaModel(config, add_pooling_layer=False)
        classifier_dropout = (
            config.classifier_dropout if config.classifier_dropout is not None else config.hidden_dropout_prob
        )
        self.dropout = nn.Dropout(classifier_dropout)
        self.classifier = nn.Linear(config.hidden_size + 3, config.num_labels)

        # Initialize weights and apply final processing
        self.post_init()

    @add_start_docstrings_to_model_forward(ROBERTA_INPUTS_DOCSTRING.format("batch_size, sequence_length"))
    @add_code_sample_docstrings(
        checkpoint="Jean-Baptiste/roberta-large-ner-english",
        output_type=TokenClassifierOutput,
        config_class=_CONFIG_FOR_DOC,
        expected_output="['O', 'ORG', 'ORG', 'O', 'O', 'O', 'O', 'O', 'LOC', 'O', 'LOC', 'LOC']",
        expected_loss=0.01,
    )
    def forward(
        self,
        input_ids: Optional[torch.LongTensor] = None,
        attention_mask: Optional[torch.FloatTensor] = None,
        token_type_ids: Optional[torch.LongTensor] = None,
        position_ids: Optional[torch.LongTensor] = None,
        per: Optional[torch.Tensor] = None,
        org: Optional[torch.Tensor] = None,
        loc: Optional[torch.Tensor] = None,
        head_mask: Optional[torch.FloatTensor] = None,
        inputs_embeds: Optional[torch.FloatTensor] = None,
        labels: Optional[torch.LongTensor] = None,
        output_attentions: Optional[bool] = None,
        output_hidden_states: Optional[bool] = None,
        return_dict: Optional[bool] = None,
    ) -> Union[Tuple[torch.Tensor], TokenClassifierOutput]:
        r"""
        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
            Labels for computing the token classification loss. Indices should be in `[0, ..., config.num_labels - 1]`.
        """
        return_dict = return_dict if return_dict is not None else self.config.use_return_dict

        outputs = self.roberta(
            input_ids,
            attention_mask=attention_mask,
            token_type_ids=token_type_ids,
            position_ids=position_ids,
            per=per,
            org=org,
            loc=loc,
            head_mask=head_mask,
            inputs_embeds=inputs_embeds,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

        sequence_output = outputs[0]

        sequence_output = self.dropout(sequence_output)
        logits = self.classifier(sequence_output)

        loss = None
        if labels is not None:
            # move labels to correct device to enable model parallelism
            labels = labels.to(logits.device)
            loss_fct = nn.CrossEntropyLoss()
            loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))

        if not return_dict:
            output = (logits,) + outputs[2:]
            return ((loss,) + output) if loss is not None else output

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