Upload 5 files

Dockerfile

@@ -0,0 +1,28 @@
+FROM python:3.9-slim
+
+# Install curl and Ollama
+RUN apt-get update && apt-get install -y curl && \
+    curl -fsSL https://ollama.ai/install.sh | sh && \
+    apt-get clean && rm -rf /var/lib/apt/lists/*
+
+# Set up user and environment
+RUN useradd -m -u 1000 user
+USER user
+ENV HOME=/home/user \
+    PATH="/home/user/.local/bin:$PATH"
+
+WORKDIR $HOME/app
+
+COPY --chown=user requirements.txt .
+RUN pip install --no-cache-dir --upgrade -r requirements.txt
+
+COPY --chown=user . .
+
+# Make the start script executable
+RUN chmod +x start.sh
+
+# Expose ports for FastAPI (7860) and Streamlit (8501)
+EXPOSE 7860
+EXPOSE 8501
+
+CMD ["./start.sh"]

main.py

@@ -0,0 +1,44 @@
+import os
+import logging
+from fastapi import FastAPI, HTTPException
+from fastapi.responses import StreamingResponse
+from pydantic import BaseModel
+from langchain_community.llms import Ollama
+from langchain.callbacks.manager import CallbackManager
+from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+app = FastAPI()
+MODEL_NAME = 'krishna_choudhary/AI_Assistant_Chatbot'
+
+def get_llm():
+    callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
+    return Ollama(model=MODEL_NAME, callback_manager=callback_manager)
+
+class Question(BaseModel):
+    text: str
+
+@app.get("/")
+def read_root():
+    return {"Hello": f"Welcome to {MODEL_NAME} FastAPI"}
+
+@app.post("/ask")
+async def ask_question(question: Question):
+    try:
+        llm = get_llm()
+        response = llm.invoke(question.text)
+        return {"response": response}
+    except Exception as e:
+        logger.error(f"Error processing question: {e}")
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@app.on_event("startup")
+async def startup_event():
+    logger.info(f"Starting up with model: {MODEL_NAME}")
+
+@app.on_event("shutdown")
+async def shutdown_event():
+    logger.info("Shutting down")

requirements.txt

@@ -0,0 +1,7 @@
+fastapi
+uvicorn
+langchain
+langchain_community
+ollama
+streamlit
+requests

start.sh

@@ -0,0 +1,66 @@
+#!/bin/bash
+
+# Set environment variables for optimization
+export OMP_NUM_THREADS=4
+export MKL_NUM_THREADS=4
+export CUDA_VISIBLE_DEVICES=0
+
+echo "--- Starting Ollama, FastAPI, and Streamlit ---"
+
+# Start Ollama in the background
+echo "Starting Ollama server..."
+ollama serve &
+
+# Pull the model if not already present
+MODEL_TO_PULL="krishna_choudhary/AI_Assistant_Chatbot"
+if ! ollama list | grep -q "$MODEL_TO_PULL"; then
+    echo "Pulling Ollama model: $MODEL_TO_PULL"
+    ollama pull "$MODEL_TO_PULL"
+else
+    echo "Ollama model $MODEL_TO_PULL already present."
+fi
+
+# Wait for Ollama to start up
+max_attempts=90 # Increased attempts as model pulling can take time on first run
+attempt=0
+echo "Waiting for Ollama to start (max $max_attempts seconds)..."
+while ! curl -s http://localhost:11434/api/tags >/dev/null; do
+    sleep 1
+    attempt=$((attempt + 1))
+    if [ $attempt -eq $max_attempts ]; then
+        echo "Ollama failed to start within $((max_attempts)) seconds. Exiting."
+        exit 1
+    fi
+done
+echo "Ollama is ready."
+
+# --- DEBUGGING: List files in current directory ---
+echo "--- Files in current directory ($PWD): ---"
+ls -l
+echo "-------------------------------------------"
+
+# Start the FastAPI server in the background
+echo "Starting FastAPI server..."
+uvicorn main:app --host 0.0.0.0 --port 7860 --workers 1 --limit-concurrency 20 &
+
+# Store the PID of the FastAPI server
+FASTAPI_PID=$!
+echo "FastAPI server started with PID: $FASTAPI_PID"
+
+# Wait a moment for FastAPI to start (optional, but good practice)
+sleep 5
+
+# Start the Streamlit server in the background
+echo "Starting Streamlit app..."
+# Use 'python -m streamlit' for robustness in Docker environments
+python -m streamlit run streamlit_app.py --server.port 8501 --server.address 0.0.0.0 &
+
+# Store the PID of the Streamlit server
+STREAMLIT_PID=$!
+echo "Streamlit app started with PID: $STREAMLIT_PID"
+
+echo "All services initiated. Keeping container alive..."
+
+# Keep the script running indefinitely, so the Docker container doesn't exit.
+# This is more robust than `wait -n` if one background process exits unexpectedly.
+tail -f /dev/null

streamlit_app.py

@@ -0,0 +1,69 @@
+import streamlit as st
+import requests
+import json
+
+# FastAPI configuration
+FASTAPI_URL = "http://localhost:7860/ask"
+
+st.set_page_config(page_title="Ollama AI Assistant", page_icon="🤖", layout="wide")
+
+# --- Session state for chat history ---
+if 'chat_history' not in st.session_state:
+    st.session_state.chat_history = [
+        {"role": "assistant", "message": "Hello! How can I assist you today?"}
+    ]
+
+# --- App Header ---
+st.title("🤖 Ollama AI Assistant")
+st.caption("Start chatting with our AI assistant. Type your message below and press send.")
+
+# --- Chat Display ---
+st.markdown("---")
+for chat in st.session_state.chat_history:
+    if chat["role"] == "assistant":
+        with st.chat_message("assistant", avatar="🤖"):
+            st.write(chat["message"])
+    else:
+        with st.chat_message("user"):
+            st.write(chat["message"])
+
+# --- Input Area ---
+with st.form("chat_form", clear_on_submit=True):
+    user_prompt = st.text_area(
+        "Type your message here...",
+        height=100,
+        placeholder="e.g., Explain quantum computing in simple terms.",
+        label_visibility="collapsed",
+        key="user_input_text_area"
+    )
+    submitted = st.form_submit_button("Send")
+
+    if submitted and user_prompt:
+        st.session_state.chat_history.append({"role": "user", "message": user_prompt})
+        with st.chat_message("assistant", avatar="🤖"):
+            st.write("Thinking...")
+
+        try:
+            payload = {"text": user_prompt}
+            headers = {"Content-Type": "application/json"}
+            response = requests.post(FASTAPI_URL, data=json.dumps(payload), headers=headers)
+
+            if response.status_code == 200:
+                llm_response = response.json().get("response", "No response received.")
+            else:
+                llm_response = f"Error: FastAPI server returned {response.status_code}. Details: {response.text}"
+
+        except requests.exceptions.ConnectionError:
+            llm_response = f"Error: Cannot connect to the FastAPI server at {FASTAPI_URL}."
+        except Exception as e:
+            llm_response = f"Unexpected error: {e}"
+
+        st.session_state.chat_history.append({"role": "assistant", "message": llm_response})
+        st.rerun()
+
+    elif submitted and not user_prompt:
+        st.warning("Please enter a prompt before clicking 'Send'.")
+
+# --- Footer ---
+st.markdown("---")
+st.caption("Powered by Ollama, FastAPI, and Streamlit.")