"""
ML-based profanity and toxicity detection module.
This module provides optional ML-based detection using:
- detoxify (transformer-based, more accurate)
- profanity-check (SVM-based, faster)
Install dependencies:
# For transformer-based detection (slower, more accurate)
pip install detoxify torch
# For lightweight detection (faster, simpler)
pip install profanity-check
# Or install with extras
pip install glin-profanity[ml] # Both
pip install glin-profanity[ml-transformer] # Just transformer
pip install glin-profanity[ml-lightweight] # Just lightweight
Example:
>>> # Standalone lightweight detector (fast)
>>> from glin_profanity.ml import LightweightDetector
>>>
>>> detector = LightweightDetector()
>>> result = detector.analyze("some text")
>>> print(result["is_toxic"])
>>>
>>> # Standalone transformer detector (accurate)
>>> from glin_profanity.ml import ToxicityDetector
>>>
>>> detector = ToxicityDetector({"model_type": "original"})
>>> result = detector.analyze("some text")
>>>
>>> # Hybrid filter (combines rules + ML)
>>> from glin_profanity.ml import HybridFilter
>>>
>>> filter = HybridFilter({
... "languages": ["english"],
... "detect_leetspeak": True,
... "enable_ml": True,
... "ml_type": "lightweight",
... })
>>> result = filter.check_profanity_hybrid("some text")
"""
from .hybrid import HybridFilter, HybridFilterConfig
from .lightweight import LightweightDetector
from .toxicity import ToxicityDetector
from .types import (
CombinationMode,
HybridAnalysisResult,
MLAnalysisResult,
MLDetectorConfig,
ToxicityLabel,
ToxicityPrediction,
)
__all__ = [
# Detectors
"ToxicityDetector",
"LightweightDetector",
"HybridFilter",
# Config types
"HybridFilterConfig",
"MLDetectorConfig",
# Result types
"MLAnalysisResult",
"HybridAnalysisResult",
"ToxicityLabel",
"ToxicityPrediction",
"CombinationMode",
]
