- RAG & SPIN Specialist improvements

This commit is contained in:
Josako
2025-04-22 13:49:38 +02:00
parent 4bf12db142
commit 9652d0bff9
12 changed files with 24 additions and 41 deletions

View File

@@ -1,4 +1,4 @@
from typing import Dict, Any
from typing import Dict, Any, Optional
from flask import current_app
from pydantic import BaseModel, Field, model_validator
@@ -11,6 +11,7 @@ class RetrieverMetadata(BaseModel):
document_id: int = Field(..., description="ID of the source document")
version_id: int = Field(..., description="Version ID of the source document")
document_name: str = Field(..., description="Name of the source document")
url: Optional[str] = Field(..., description="URL of the source document")
user_metadata: Dict[str, Any] = Field(
default_factory=dict, # This will use an empty dict if None is provided
description="User-defined metadata"

View File

@@ -97,6 +97,7 @@ class StandardRAGRetriever(BaseRetriever):
query_obj = (
db.session.query(
db_class,
DocumentVersion.url,
(1 - db_class.embedding.cosine_distance(query_embedding)).label('similarity')
)
.join(DocumentVersion, db_class.doc_vers_id == DocumentVersion.id)
@@ -116,7 +117,7 @@ class StandardRAGRetriever(BaseRetriever):
# Transform results into standard format
processed_results = []
for doc, similarity in results:
for doc, url, similarity in results:
# Parse user_metadata to ensure it's a dictionary
user_metadata = self._parse_metadata(doc.document_version.user_metadata)
processed_results.append(
@@ -128,6 +129,7 @@ class StandardRAGRetriever(BaseRetriever):
document_id=doc.document_version.doc_id,
version_id=doc.document_version.id,
document_name=doc.document_version.document.name,
url=url or "",
user_metadata=user_metadata,
)
)