Upload 2 files

app.py

@@ -0,0 +1,90 @@
+import os
+import threading
+import time
+import subprocess
+
+print("Expanding user path for Ollama")
+OLLAMA = os.path.expanduser("~/ollama")
+
+print("Checking if Ollama exists at the path")
+if not os.path.exists(OLLAMA):
+    print("Ollama not found, downloading it")
+    subprocess.run("curl -L https://ollama.com/download/ollama-linux-amd64 -o ~/ollama", shell=True)
+    os.chmod(OLLAMA, 0o755)
+
+def ollama_service_thread():
+    print("Starting Ollama service thread")
+    subprocess.run("~/ollama serve", shell=True)
+
+print("Creating and starting Ollama service thread")
+OLLAMA_SERVICE_THREAD = threading.Thread(target=ollama_service_thread)
+OLLAMA_SERVICE_THREAD.start()
+
+print("Giving Ollama serve a moment to start")
+time.sleep(10)
+
+print("Setting model to 'gemma2'")
+model = "gemma2"
+
+print(f"Pulling model {model}")
+subprocess.run(f"~/ollama pull {model}", shell=True)
+
+################################################
+################################################
+import copy
+import gradio as gr
+from ollama import Client
+
+print("Initializing Ollama client")
+client = Client(host='http://localhost:11434', timeout=120)
+
+print("Getting Hugging Face token and model ID from environment variables")
+HF_TOKEN = os.environ.get("HF_TOKEN", None)
+MODEL_ID = os.environ.get("MODEL_ID", "google/gemma-2-9b-it")
+MODEL_NAME = MODEL_ID.split("/")[-1]
+
+print("Setting up title and description for Gradio interface")
+TITLE = "<h1><center>ollama-Chat</center></h1>"
+DESCRIPTION = f"""
+<h3>MODEL: <a href="https://hf.co/{MODEL_ID}">{MODEL_NAME}</a></h3>
+<p>Running on Ollama backend.</p>
+"""
+
+
+CSS = """
+.duplicate-button {
+    margin: auto !important;
+    color: white !important;
+    background: black !important;
+    border-radius: 100vh !important;
+}
+h3 {
+    text-align: center;
+}
+"""
+import streamlit as st
+from llama_index.llms.ollama import Ollama
+
+# Initialize the Ollama model
+llm = Ollama(model="llama3", request_timeout=120.0)
+
+st.title("Paul Graham Information Fetcher")
+
+# Text input for the query
+query = st.text_input("Enter your query:", value="Who is Paul Graham?")
+
+# Button to trigger the API call
+if st.button("Get Response"):
+    with st.spinner("Fetching response..."):
+        try:
+            # Fetch the response from the model
+            resp = llm.complete(query)
+            # Display the response
+            st.success("Response fetched successfully!")
+            st.write(resp)
+        except Exception as e:
+            st.error(f"An error occurred: {e}")
+
+# Run the Streamlit app
+if __name__ == "__main__":
+    st.run()

requirements.txt

@@ -0,0 +1,3 @@
+ollama
+streamlit
+llama_index.llms.ollama