File size: 3,731 Bytes
74c680c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
from __future__ import annotations

import json
import os
from typing import Any, Dict, List

import torch
from dotenv import load_dotenv
from huggingface_hub import login
from transformers import AutoModelForCausalLM, AutoTokenizer

load_dotenv()

HF_TOKEN = os.getenv("HF_TOKEN", "")
MODEL_NAME = os.getenv("MODEL_NAME", "meta-llama/Llama-3.2-3B-Instruct")
MODEL_ID = MODEL_NAME

_tokenizer = None
_model = None


def _get_env(key: str, default=None, required: bool = False, cast=None):
    value = os.getenv(key, default)

    if required and value is None:
        raise ValueError(f"Missing required env var: {key}")

    if cast and value is not None:
        try:
            value = cast(value)
        except Exception as e:
            raise ValueError(f"Failed to cast env var {key}: {e}")

    return value


def _extract_json(text: str) -> Dict[str, Any]:
    if not text:
        return {"label": "error", "reason": "empty model output", "raw": text}

    start = text.find("{")
    end = text.rfind("}")
    if start != -1 and end != -1 and end > start:
        candidate = text[start:end + 1]
        try:
            parsed = json.loads(candidate)
            if isinstance(parsed, dict):
                return parsed
        except Exception:
            pass

    return {"label": "error", "reason": "could not parse JSON", "raw": text}


def load_model():
    global _tokenizer, _model

    if _tokenizer is not None and _model is not None:
        return _tokenizer, _model

    if HF_TOKEN:
        try:
            login(token=HF_TOKEN)
        except Exception as e:
            print(f"HF login warning: {e}")

    kwargs = {}
    if HF_TOKEN:
        kwargs["token"] = HF_TOKEN

    print(f"Loading tokenizer for {MODEL_ID}...")
    _tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, **kwargs)

    if _tokenizer.pad_token_id is None:
        _tokenizer.pad_token = _tokenizer.eos_token

    dtype = torch.float16 if torch.cuda.is_available() else torch.float32

    print(f"Loading model for {MODEL_ID}...")
    _model = AutoModelForCausalLM.from_pretrained(
        MODEL_ID,
        torch_dtype=dtype,
        device_map="auto" if torch.cuda.is_available() else None,
        **kwargs,
    )
    _model.eval()

    print("Model loaded.")
    return _tokenizer, _model


def relation_model_wrapper(messages: List[Dict[str, str]]) -> Dict[str, Any]:
    """
    Expects chat-style messages:
    [
        {"role": "system", "content": "..."},
        {"role": "user", "content": "..."},
    ]
    """
    tokenizer, model = load_model()

    try:
        prompt = tokenizer.apply_chat_template(
            messages,
            tokenize=False,
            add_generation_prompt=True,
        )

        inputs = tokenizer(
            prompt,
            return_tensors="pt",
            truncation=True,
            max_length=4096,
        )

        device = model.device if hasattr(model, "device") else next(model.parameters()).device
        inputs = {k: v.to(device) for k, v in inputs.items()}

        with torch.no_grad():
            output = model.generate(
                **inputs,
                max_new_tokens=160,
                do_sample=False,
                pad_token_id=tokenizer.eos_token_id,
                eos_token_id=tokenizer.eos_token_id,
            )

        generated = tokenizer.decode(
            output[0][inputs["input_ids"].shape[1]:],
            skip_special_tokens=True,
        )

        parsed = _extract_json(generated)
        parsed.setdefault("raw", generated)
        return parsed

    except Exception as e:
        return {
            "label": "error",
            "reason": f"model wrapper failure: {str(e)}",
            "raw": "",
        }