Reducing startup time/cost

chunker.py

@@ -8,6 +8,7 @@ from chromadb.utils.embedding_functions.openai_embedding_function import (
     OpenAIEmbeddingFunction,
 )
 from dotenv import load_dotenv
+from llm import LLMClient
 
 
 USE_OPENAI = os.getenv("OPENAI_API_KEY") != None
@@ -91,9 +92,10 @@ class Chunker:
 
     def __init__(self, collection) -> None:
         self.collection = collection
+        self.llm_client = LLMClient()
 
     def embedding_fx(self, inputs):
-        if USE_OPENAI:
+        if self.llm_client.PROVIDER == "openai":
             openai_embedding_fx = OpenAIEmbeddingFunction(
                 api_key=os.getenv("OPENAI_API_KEY"),
                 model_name="text-embedding-3-small",

llm.py

@@ -16,7 +16,7 @@ class LLMClient:
             self.ollama_client = Client(
                 host=os.getenv("OLLAMA_URL", "http://localhost:11434")
             )
-            client.chat(
+            self.ollama_client.chat(
                 model="gemma3:4b", messages=[{"role": "system", "content": "test"}]
             )
             self.PROVIDER = "ollama"
@@ -35,9 +35,16 @@ class LLMClient:
         if self.PROVIDER == "ollama":
             response = self.ollama_client.chat(
                 model="gemma3:4b",
-                prompt=prompt,
+                messages=[
+                    {
+                        "role": "system",
+                        "content": system_prompt,
+                    },
+                    {"role": "user", "content": prompt},
+                ],
             )
-            output = response["response"]
+            print(response)
+            output = response.message.content
         elif self.PROVIDER == "openai":
             response = self.openai_client.responses.create(
                 model="gpt-4o-mini",
@@ -51,6 +58,8 @@ class LLMClient:
             )
             output = response.output_text
 
+        return output
+
 
 if __name__ == "__main__":
     client = Client()

main.py

@@ -1,6 +1,7 @@
 import datetime
 import logging
 import os
+import sqlite3
 from typing import Any, Union
 
 import argparse
@@ -15,6 +16,7 @@ from query import QueryGenerator
 from cleaner import pdf_to_image, summarize_pdf_image
 from llm import LLMClient
 
+
 from dotenv import load_dotenv
 
 load_dotenv()
@@ -25,7 +27,7 @@ USE_OPENAI = os.getenv("OPENAI_API_KEY") != None
 ollama_client = ollama.Client(host=os.getenv("OLLAMA_URL", "http://localhost:11434"))
 
 client = chromadb.PersistentClient(path=os.getenv("CHROMADB_PATH", ""))
-simba_docs = client.get_or_create_collection(name="simba_docs2")
+simba_docs = client.get_or_create_collection(name="simba_docs3")
 feline_vet_lookup = client.get_or_create_collection(name="feline_vet_lookup")
 
 parser = argparse.ArgumentParser(
@@ -76,6 +78,9 @@ def chunk_data(docs: list[dict[str, Union[str, Any]]], collection):
 
     print(f"chunking {len(docs)} documents")
     texts: list[str] = [doc["content"] for doc in docs]
+    with sqlite3.connect("visited.db") as conn: 
+        to_insert = []
+        c = conn.cursor()
         for index, text in enumerate(texts):
             metadata = {
                 "created_date": date_to_epoch(docs[index]["created_date"]),
@@ -85,6 +90,10 @@ def chunk_data(docs: list[dict[str, Union[str, Any]]], collection):
                 document=text,
                 metadata=metadata,
             )
+            to_insert.append((docs[index]["id"],))
+
+        c.executemany("INSERT INTO indexed_documents (paperless_id) values (?)", to_insert)
+
 
 
 def chunk_text(texts: list[str], collection):
@@ -160,6 +169,18 @@ def consult_simba_oracle(input: str):
         collection=simba_docs,
     )
 
+def filter_indexed_files(docs):
+    with sqlite3.connect("visited.db") as conn:
+        c = conn.cursor()
+        c.execute("CREATE TABLE IF NOT EXISTS indexed_documents (id INTEGER PRIMARY KEY AUTOINCREMENT, paperless_id INTEGER)")
+        c.execute("SELECT paperless_id FROM indexed_documents")
+        rows = c.fetchall()
+        conn.commit()
+
+    visited = {row[0] for row in rows}
+    return [doc for doc in docs if doc["id"] not in visited]
+
+
 
 if __name__ == "__main__":
     args = parser.parse_args()
@@ -167,6 +188,7 @@ if __name__ == "__main__":
         print("Fetching documents from Paperless-NGX")
         ppngx = PaperlessNGXService()
         docs = ppngx.get_data()
+        docs = filter_indexed_files(docs)
         print(f"Fetched {len(docs)} documents")
         #
         print("Chunking documents now ...")