modeling_ocrqa.py

import torch
import torch.nn as nn
from transformers import PreTrainedModel
import logging
import floret
import os
from huggingface_hub import hf_hub_download
from .configuration_lang import ImpressoConfig

logger = logging.getLogger(__name__)


class QAAssessmentModel(PreTrainedModel):
    config_class = ImpressoConfig

    def __init__(self, config):
        super().__init__(config)
        self.config = config

        # Dummy for device checking
        self.dummy_param = nn.Parameter(torch.zeros(1))
        bin_filename = self.config.config.filename

        # Check if the file is already present locally, else download it
        if not os.path.exists(bin_filename):
            # print(f"{bin_filename} not found locally, downloading from Hugging Face hub...")
            bin_filename = hf_hub_download(repo_id=self.config.config._name_or_path, 
                                       filename=bin_filename)

        # Load floret model using the full path
        self.model_floret = floret.load_model(bin_filename)

    def forward(self, input_ids, **kwargs):
        if isinstance(input_ids, str):
            # If the input is a single string, make it a list for floret
            texts = [input_ids]
        elif isinstance(input_ids, list) and all(isinstance(t, str) for t in input_ids):
            texts = input_ids
        else:
            raise ValueError(f"Unexpected input type: {type(input_ids)}")

        predictions, probabilities = self.model_floret.predict(texts, k=1)
        return (
            predictions,
            probabilities,
        )

    @property
    def device(self):
        return next(self.parameters()).device

    @classmethod
    def from_pretrained(cls, *args, **kwargs):
        # print("Ignoring weights and using custom initialization.")
        # Manually create the config
        config = ImpressoConfig(**kwargs)
        # Pass the manually created config to the class
        model = cls(config)
        return model