ML-Training-Module/config.py at main · Noface07/ML-Training-Module · GitHub

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
"""
Application configuration and use-case registry.

Settings are loaded from environment variables / .env file via pydantic-settings.
The USE_CASE_REGISTRY is the single source of truth for mapping use-case names
to model types, output tags, and training requirements.
"""

from __future__ import annotations

from pydantic_settings import BaseSettings


class Settings(BaseSettings):
    """Central configuration for the ML Platform."""

    # ── Paths ──────────────────────────────────────────────────────────
    ARTIFACTS_DIR: str = "artifacts"
    DATASETS_DIR: str = "datasets"
    LOGS_DIR: str = "logs"
    DB_URL: str = "sqlite:///./ml_platform.db"

    # ── Ingest defaults ────────────────────────────────────────────────
    DEFAULT_TIMESTAMP_COL: str = "timestamp"
    DEFAULT_FORWARD_FILL_LIMIT: int = 5
    DEFAULT_BACKWARD_FILL: bool = True
    DEFAULT_TRAIN_SPLIT: float = 0.8
    DEFAULT_CV_FOLDS: int = 5

    # ── SSE ─────────────────────────────────────────────────────────────
    SSE_POLL_INTERVAL_MS: int = 200

    model_config = {"env_file": ".env", "extra": "ignore"}


settings = Settings()


# ── Use-case registry ──────────────────────────────────────────────────
# Maps every supported use-case to its model type, output tag, and
# whether it needs a full training run (as opposed to a statistical /
# rule-based pipeline).

USE_CASE_REGISTRY: dict[str, dict] = {
    "failure_prediction": {
        "model_type": "xgboost_clf",
        "output_tag": "risk_score",
        "requires_training": True,
    },
    "risk_scoring": {
        "model_type": "xgboost_clf",
        "output_tag": "risk_score",
        "requires_training": True,
    },
    "rul": {
        "model_type": "xgboost_reg",
        "output_tag": "rul_hours",
        "requires_training": True,
    },
    "next_interval": {
        "model_type": "xgboost_reg",
        "output_tag": "next_value",
        "requires_training": True,
    },
    "anomaly_multivariate": {
        "model_type": "isolation_forest",
        "output_tag": "anomaly_score",
        "requires_training": True,
    },
    "kpi_prediction": {
        "model_type": "xgboost_reg",
        "output_tag": "kpi_value",
        "requires_training": True,
    },
    "anomaly_univariate": {
        "model_type": "statistical",
        "output_tag": "anomaly_score",
        "requires_training": False,
    },
    "adaptive_threshold": {
        "model_type": "statistical",
        "output_tag": "dynamic_threshold",
        "requires_training": False,
    },
    "early_warning": {
        "model_type": "statistical",
        "output_tag": "warning_score",
        "requires_training": False,
    },
    "health_index": {
        "model_type": "statistical",
        "output_tag": "health_score",
        "requires_training": False,
    },
    "drift_detection": {
        "model_type": "statistical",
        "output_tag": "drift_score",
        "requires_training": False,
    },
    "pattern_detection": {
        "model_type": "statistical",
        "output_tag": "pattern_flag",
        "requires_training": False,
    },
    "data_quality": {
        "model_type": "statistical",
        "output_tag": "data_quality_score",
        "requires_training": False,
    },
}