Lars/mindnet
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 21 Wiki Activity

scripts/setup_mindnet_collections.py aktualisiert
All checks were successful
Deploy mindnet to llm-node / deploy (push) Successful in 2s
Details

Browse Source
This commit is contained in:
Lars 2025-09-01 15:00:21 +02:00
parent a7f3fc5784
commit 12043b7752
1 changed files with 7 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
scripts/setup_mindnet_collections.py
View File

@ -1,12 +1,12 @@
#!/usr/bin/env python3 #!/usr/bin/env python3
""" """
Erzeugt Qdrant-Collections für das mindnet-Projekt: Richtet die Qdrant-Collections für das mindnet-Projekt ein.
- mindnet_chunks : semantische Suche über Markdown-Text-Chunks (384/Cosine) - mindnet_chunks : semantische Suche über Text-Chunks (384/Cosine)
- mindnet_notes : 1 Punkt pro Notiz (Metadaten, optional Titel-Embedding) - mindnet_notes : 1 Punkt pro Notiz (optional Titel-Embedding)
- mindnet_edges : explizite Link-Kanten (Dummy-Vektor size=1; Filter über Payload) - mindnet_edges : explizite Link-Kanten (Dummy-Vektor size=1; Filter über Payload)
Idempotent: legt nur an, wenn nicht vorhanden. Legt sinnvolle Payload-Indizes an. Idempotent: legt nur an, wenn nicht vorhanden.
""" """
import os import os
@ -55,6 +55,7 @@ def main():
ap.add_argument("--distance", default="Cosine", choices=["Cosine", "Euclid", "Dot"], help="Distanzmetrik") ap.add_argument("--distance", default="Cosine", choices=["Cosine", "Euclid", "Dot"], help="Distanzmetrik")
args = ap.parse_args() args = ap.parse_args()
# Qdrant-URL überschreiben, falls per Argument gesetzt
global DEFAULT_QDRANT_URL global DEFAULT_QDRANT_URL
DEFAULT_QDRANT_URL = args.qdrant_url DEFAULT_QDRANT_URL = args.qdrant_url
@ -67,21 +68,18 @@ def main():
create_collection(notes, size=args.dim, distance=args.distance) create_collection(notes, size=args.dim, distance=args.distance)
create_collection(edges, size=1, distance=args.distance) # Dummy-Vektor create_collection(edges, size=1, distance=args.distance) # Dummy-Vektor
# 2) Indizes definieren # 2) Indizes
# mindnet_chunks: häufige Filter + Volltext
for f in ["note_id", "Status", "Typ", "title", "path"]: for f in ["note_id", "Status", "Typ", "title", "path"]:
create_keyword_index(chunks, f) create_keyword_index(chunks, f)
for f in ["tags", "Rolle", "links"]: for f in ["tags", "Rolle", "links"]:
create_keyword_index(chunks, f) create_keyword_index(chunks, f)
create_text_index(chunks, "text") # Wort-/Phrasensuche create_text_index(chunks, "text")
# mindnet_notes: Metadaten der Notizen
for f in ["note_id", "title", "path", "Typ", "Status"]: for f in ["note_id", "title", "path", "Typ", "Status"]:
create_keyword_index(notes, f) create_keyword_index(notes, f)
for f in ["tags", "Rolle"]: for f in ["tags", "Rolle"]:
create_keyword_index(notes, f) create_keyword_index(notes, f)
# mindnet_edges: Graph/Kanten (Filter-only)
for f in ["src_note_id", "dst_note_id", "src_chunk_id", "dst_chunk_id", "link_text", "relation"]: for f in ["src_note_id", "dst_note_id", "src_chunk_id", "dst_chunk_id", "link_text", "relation"]:
create_keyword_index(edges, f) create_keyword_index(edges, f)