Add Phase 13 internal accounting model

2026-08-22 21:52:44 +05:30
parent ef2b0e1f83
commit 0c0457aaca
15 changed files with 1171 additions and 3 deletions
@@ -0,0 +1,78 @@
+from __future__ import annotations
+
+from datetime import datetime, timezone
+
+from sqlalchemy import Boolean, DateTime, Float, ForeignKey, Integer, String, Text
+from sqlalchemy.orm import Mapped, mapped_column
+
+from app.core.db.common import CommonBase
+
+
+class AccountingInternalModel(CommonBase):
+    __tablename__ = "accounting_internal_models"
+
+    id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
+    tenant_id: Mapped[int] = mapped_column(ForeignKey("tenants.id", ondelete="CASCADE"), nullable=False, index=True)
+
+    version_label: Mapped[str] = mapped_column(String(80), nullable=False, index=True)
+    algorithm: Mapped[str] = mapped_column(String(80), nullable=False, default="multinomial_nb_v1")
+    status: Mapped[str] = mapped_column(String(30), nullable=False, default="shadow", index=True)
+    is_active: Mapped[bool] = mapped_column(Boolean, nullable=False, default=False, index=True)
+
+    training_examples: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
+    validation_examples: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
+    class_count: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
+    vocabulary_size: Mapped[int] = mapped_column(Integer, nullable=False, default=0)
+
+    validation_accuracy: Mapped[float] = mapped_column(Float, nullable=False, default=0)
+    validation_macro_recall: Mapped[float] = mapped_column(Float, nullable=False, default=0)
+    validation_top2_accuracy: Mapped[float] = mapped_column(Float, nullable=False, default=0)
+
+    model_json: Mapped[str] = mapped_column(Text, nullable=False)
+    metrics_json: Mapped[str | None] = mapped_column(Text, nullable=True)
+    training_summary_json: Mapped[str | None] = mapped_column(Text, nullable=True)
+
+    trained_by_user_id: Mapped[int | None] = mapped_column(
+        ForeignKey("users.id", ondelete="SET NULL"), nullable=True
+    )
+    trained_at_utc: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, default=lambda: datetime.now(timezone.utc), index=True
+    )
+
+
+class AccountingInternalPrediction(CommonBase):
+    __tablename__ = "accounting_internal_predictions"
+
+    id: Mapped[int] = mapped_column(Integer, primary_key=True, autoincrement=True)
+    tenant_id: Mapped[int] = mapped_column(ForeignKey("tenants.id", ondelete="CASCADE"), nullable=False, index=True)
+    client_id: Mapped[int] = mapped_column(ForeignKey("clients.id", ondelete="CASCADE"), nullable=False, index=True)
+    model_id: Mapped[int] = mapped_column(
+        ForeignKey("accounting_internal_models.id", ondelete="CASCADE"), nullable=False, index=True
+    )
+
+    source_type: Mapped[str] = mapped_column(String(30), nullable=False, index=True)
+    source_record_id: Mapped[int] = mapped_column(Integer, nullable=False, index=True)
+    source_fingerprint: Mapped[str] = mapped_column(String(80), nullable=False, default="", index=True)
+
+    predicted_nature_id: Mapped[int | None] = mapped_column(
+        ForeignKey("accounting_natures.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    predicted_probability: Mapped[float] = mapped_column(Float, nullable=False, default=0)
+    top2_json: Mapped[str | None] = mapped_column(Text, nullable=True)
+    explanation_json: Mapped[str | None] = mapped_column(Text, nullable=True)
+
+    shadow_mode: Mapped[bool] = mapped_column(Boolean, nullable=False, default=True, index=True)
+    applied_to_source: Mapped[bool] = mapped_column(Boolean, nullable=False, default=False, index=True)
+
+    final_nature_id: Mapped[int | None] = mapped_column(
+        ForeignKey("accounting_natures.id", ondelete="SET NULL"), nullable=True, index=True
+    )
+    prediction_correct: Mapped[bool | None] = mapped_column(Boolean, nullable=True, index=True)
+    reviewed_by_user_id: Mapped[int | None] = mapped_column(
+        ForeignKey("users.id", ondelete="SET NULL"), nullable=True
+    )
+    reviewed_at_utc: Mapped[datetime | None] = mapped_column(DateTime(timezone=True), nullable=True)
+
+    created_at_utc: Mapped[datetime] = mapped_column(
+        DateTime(timezone=True), nullable=False, default=lambda: datetime.now(timezone.utc), index=True
+    )