File size: 1,576 Bytes
cb7d23f
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e32e347
cb7d23f
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
# Necessary imports
import os
import sys
from dotenv import load_dotenv
from typing import Any
import torch
from transformers import AutoModel, AutoTokenizer

# Local imports
from src.logger import logging
from src.exception import CustomExceptionHandling


# Load the Environment Variables from .env file
load_dotenv()

# Access token for using the model
access_token = os.environ.get("ACCESS_TOKEN")


def load_model_and_tokenizer(model_name: str, device: str) -> Any:
    """
    Load the model and tokenizer.

    Args:
        - model_name (str): The name of the model to load.
        - device (str): The device to load the model onto.

    Returns:
        - model: The loaded model.
        - tokenizer: The loaded tokenizer.
    """
    try:
        # Load the model and tokenizer
        model = AutoModel.from_pretrained(
            model_name,
            trust_remote_code=True,
            attn_implementation="sdpa",
            torch_dtype=torch.bfloat16,
            token=access_token
        )
        model = model.to(device=device)
        tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True, token=access_token)
        model.eval()

        # Log the successful loading of the model and tokenizer
        logging.info("Model and tokenizer loaded successfully.")

        # Return the model and tokenizer
        return model, tokenizer

    # Handle exceptions that may occur during model and tokenizer loading
    except Exception as e:
        # Custom exception handling
        raise CustomExceptionHandling(e, sys) from e