# models.py from sqlalchemy import Column, Integer, String, DateTime, Float, Boolean, Text from datetime import datetime from database import Base class User(Base): __tablename__ = "users" id = Column(Integer, primary_key=True, index=True) email = Column(String, unique=True, index=True, nullable=False) password_hash = Column(String, nullable=False) created_at = Column(DateTime, default=datetime.utcnow) # SaaS fields api_key = Column(String, unique=True, index=True, nullable=True) plan = Column(String, default="free") usage_count = Column(Integer, default=0) # Tracks the last time the user made an authenticated API request. # Useful for analytics, stale account cleanup, and pro dashboard UX. last_used = Column(DateTime, nullable=True, default=None) class PredictionLog(Base): """ Stores every /predict response for drift monitoring and analytics. Used by DriftMonitor to compute rolling confidence stats. """ __tablename__ = "prediction_logs" id = Column(Integer, primary_key=True, index=True) user_id = Column(Integer, nullable=True, index=True) text_hash = Column(String, nullable=True) # SHA256 of input (no PII) text_length = Column(Integer, nullable=True) confidence = Column(Float, nullable=False) # 0.0–1.0 toxicity probability toxic = Column(Boolean, nullable=False) severity = Column(String, nullable=True) # low/medium/high source = Column(String, nullable=True) # transformer_onnx / rules / etc. timestamp = Column(DateTime, default=datetime.utcnow, index=True) class FeedbackSample(Base): """ Active Learning feedback table. Users flag incorrect predictions (FP/FN) to build a correction queue. """ __tablename__ = "feedback_samples" id = Column(Integer, primary_key=True, index=True) user_id = Column(Integer, nullable=True, index=True) input_text = Column(Text, nullable=False) # original text predicted_toxic = Column(Boolean, nullable=False) # what model said correct_label = Column(Boolean, nullable=False) # what user says is right feedback_type = Column(String, nullable=False) # "false_positive" / "false_negative" confidence_at_time = Column(Float, nullable=True) # model confidence when flagged notes = Column(Text, nullable=True) # optional user comment abusive_words = Column(Text, nullable=True) # specific abusive words identified by user explanation = Column(Text, nullable=True) # explanation/why it is abusive reviewed = Column(Boolean, default=False) # admin reviewed flag created_at = Column(DateTime, default=datetime.utcnow, index=True)