project files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+metropolen.db filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1 @@
+__pycache__/

app.py

@@ -1,7 +1,45 @@
 import gradio as gr
+import sqlite_utils
+import llm
+import pandas as pd
+from openai import OpenAI
+import os
 
-def greet(name):
-    return "Hello " + name + "!!"
+if not os.getenv("OPENAI_API_KEY"):
+    raise ValueError("OPENAI_API_KEY must be set")
 
-demo = gr.Interface(fn=greet, inputs="text", outputs="text")
+def retrieval(question: str):
+    db = sqlite_utils.Database("metropolen.db")
+    embedding_model = llm.get_embedding_model("ada-002")
+    collection = llm.Collection("sections", db, model=embedding_model)
+
+    data = [(entry.id, entry.content, entry.score) for entry in collection.similar(question, number=3)]
+    df = pd.DataFrame(data, columns=['section_heading', 'content', 'score'])
+    documents = (df.content).str.cat(sep='\n')
+
+    return documents
+
+def make_prompt(documents, question):
+    prompt = f"""Given the context below, answer the question at the end of the text.
+    ---
+    {documents}
+    ---
+    {question}
+    """
+    return prompt
+
+def generation(prompt, api_key=None, model="gpt-4o"):
+    api_key = api_key if api_key else os.getenv('OPENAI_API_KEY')
+    client = OpenAI(api_key=api_key)
+    c = client.chat.completions.create(model=model,
+                                       messages=[{"role": "user", "content": prompt}])
+    return c.choices[0].message.content
+
+def rag(frage: str) -> str:
+    documents = retrieval(question=frage)
+    prompt = make_prompt(documents=documents, question=frage)
+    antwort = generation(prompt=prompt, api_key=os.getenv('OPENAI_API_KEY'), model="gpt-4o")
+    return antwort
+
+demo = gr.Interface(fn=rag, inputs="text", outputs="text")
 demo.launch()

metropolen.db

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60849cd41fd3ccc9aa626e37f00b450dcab9275c3b5593b06583ef3d12732080
+size 323584

requirements.txt

@@ -0,0 +1,4 @@
+sqlite_utils
+llm
+pandas
+openai