chatbot

Dockerfile

@@ -0,0 +1,12 @@
+FROM python:3.10-slim
+
+WORKDIR /app
+
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+
+COPY main.py .
+
+EXPOSE 7860
+
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]

app.env

@@ -0,0 +1 @@
+GROQ_API_KEY =  "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"

app.py

@@ -0,0 +1,92 @@
+
+
+from fastapi import FastAPI, HTTPException
+from pydantic import BaseModel
+import requests as req
+import pandas as pd
+import chromadb
+from sentence_transformers import SentenceTransformer
+from huggingface_hub import snapshot_download
+import os
+from dotenv import load_dotenv
+
+GROQ_API_KEY =  os.getenv("GROQ_API_KEY")
+GROQ_URL       = "https://api.groq.com/openai/v1/chat/completions"
+GROQ_MODEL     = "llama-3.1-8b-instant"
+CHROMA_DIR     = "chroma_db/chroma_db"
+HF_DATASET_ID  = "Mohammedelhakim/parenting-qa-vectordb"
+NUM_RESULTS    = 3
+
+
+if not os.path.exists(CHROMA_DIR):
+    print("Downloading vector database from Hugging Face Hub...")
+    snapshot_download(
+        repo_id="Mohammedelhakim/parenting-qa-vectordb",
+        repo_type="dataset",
+        local_dir="chroma_db",
+       
+)
+    print("Vector database downloaded!")
+else:
+    print("Vector database found locally, loading...")
+
+
+print("Loading embedding model...")
+embedder = SentenceTransformer("all-MiniLM-L6-v2")
+
+chroma_client = chromadb.PersistentClient(path=CHROMA_DIR)
+print(chroma_client.list_collections())
+collection    = chroma_client.get_collection("parenting_qa")
+print("Ready!")
+
+
+app = FastAPI()
+
+class ChatRequest(BaseModel):
+    message: str
+
+@app.post("/chat")
+def chat(request: ChatRequest):
+    # 1. Embed the user question
+    query_embedding = embedder.encode(request.message).tolist()
+
+    # 2. Retrieve most similar Q&As
+    results = collection.query(
+        query_embeddings=[query_embedding],
+        n_results=NUM_RESULTS,
+    )
+
+    # 3. Build context
+    context = ""
+    for doc, meta in zip(results["documents"][0], results["metadatas"][0]):
+        context += f"Q: {doc}\nA: {meta['answer']}\n\n"
+
+    # 4. Call Groq
+    messages = [
+        {"role": "system", "content": (
+            "You are a helpful and caring parenting assistant. "
+            "You answer questions from parents about their babies and children, "
+            "such as health, development, feeding, sleep, and behavior. "
+            "Give clear, reassuring, and practical answers. "
+            "If something sounds medically serious, always advise the parent to consult a doctor. "
+            "Use the following similar Q&A examples to guide your answer:\n\n"
+            + context
+        )},
+        {"role": "user", "content": request.message},
+    ]
+
+    response = req.post(
+        GROQ_URL,
+        headers={
+            "Authorization": f"Bearer {GROQ_API_KEY}",
+            "Content-Type": "application/json",
+        },
+        json={"model": GROQ_MODEL, "messages": messages},
+        timeout=30,
+    )
+
+    result = response.json()
+    if "choices" not in result:
+        raise HTTPException(status_code=500, detail=f"Groq API error: {result}")
+
+    return {"response": result["choices"][0]["message"]["content"]}

requirements.txt

@@ -0,0 +1,8 @@
+fastapi
+uvicorn
+requests
+pandas
+pyarrow
+chromadb
+sentence-transformers
+huggingface_hub