Upload app.py

app.py

@@ -1,63 +1,125 @@
-import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
-
-
-if __name__ == "__main__":
-    demo.launch()
+import os
+import threading
+import time
+import subprocess
+import socket
+
+def is_port_available(port):
+    with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
+        return s.connect_ex(('localhost', port)) != 0
+
+def kill_process_on_port(port):
+    try:
+        pid = subprocess.check_output(f"lsof -t -i:{port}", shell=True)
+        pid = pid.decode().strip()
+        if pid:
+            print(f"Killing process {pid} using port {port}")
+            subprocess.run(f"kill -9 {pid}", shell=True)
+    except subprocess.CalledProcessError:
+        print(f"No process found using port {port}")
+
+def find_available_port(start_port=11435):
+    port = start_port
+    while True:
+        with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
+            if s.connect_ex(('localhost', port)) != 0:
+                print(f"Found available port: {port}")
+                return port
+            port += 1
+
+print("Starting script...")
+
+OLLAMA = os.path.expanduser("~/ollama")
+
+if not os.path.exists(OLLAMA):
+    print("ollama path doesn't exist...")
+    subprocess.run("curl -L https://ollama.com/download/ollama-linux-amd64 -o ~/ollama", shell=True)
+    os.chmod(OLLAMA, 0o755)
+    print("Changed mod... chmod(OLLAMA, 0o755)")
+else:
+    print("ollama path exists.")
+
+# Default port is 11434
+default_port = 11434
+print(f"Default port is set to {default_port}")
+
+# Kill any process that might be using the default port
+print(f"Attempting to kill any process using port {default_port}")
+kill_process_on_port(default_port)
+print(f"{default_port} port is killed")
+
+# Verify the default port is now available
+print(f"Checking if port {default_port} is available...")
+if is_port_available(default_port):
+    available_port = default_port
+    print(f"Port {default_port} is available. Using it for the new instance.")
+else:
+    # Find another available port starting from 11435
+    print(f"Port {default_port} is not available. Finding another port...")
+    available_port = find_available_port(start_port=11435)
+    print(f"Using available port {available_port} for the new instance.")
+
+def ollama_service_thread():
+    print(f"Starting ollama service on port {available_port}...")
+    os.environ['OLLAMA_HOST'] = f'127.0.0.1:{available_port}'
+    subprocess.run("~/ollama serve", shell=True)
+    print("ollama service started.")
+
+OLLAMA_SERVICE_THREAD = threading.Thread(target=ollama_service_thread)
+OLLAMA_SERVICE_THREAD.start()
+
+print("Giving ollama serve a moment")
+time.sleep(10)
+print("Proceeding after sleep.")
+
+# Pull the model
+model = "gemma2"
+print(f"Pulling model {model}...")
+subprocess.run(f"~/ollama pull {model}", shell=True)
+print(f"Model {model} pulled.")
+
+# Import the client
+print("Importing ollama Client...")
+from ollama import Client
+print("Imported ollama Client.")
+client = Client(host=f'http://localhost:{available_port}', timeout=120)
+print(f"Client created with host: http://localhost:{available_port}")
+
+def stream_chat(message: str, history: list):
+    print("Starting stream_chat function...")
+    conversation = []
+    for prompt, answer in history:
+        conversation.extend([
+            {"role": "user", "content": prompt}, 
+            {"role": "assistant", "content": answer},
+        ])
+    conversation.append({"role": "user", "content": message})
+    print(f"Sending message to model: {message}")
+
+    response = client.chat(
+        model=model,
+        messages=conversation,
+        stream=True,
+        options={
+            'num_predict': 1024,
+            'temperature': 0.8,
+            'top_p': 0.8,
+            'top_k': 20,
+            'repeat_penalty': 1.0,
+            'low_vram': True,
+        },
+    )
+
+    buffer = ""
+    for chunk in response:
+        buffer += chunk["message"]["content"]
+        print(f"Received chunk: {chunk['message']['content']}")
+        yield buffer
+
+# Example usage
+if __name__ == "__main__":
+    print("Starting main function...")
+    history = [("Hello", "Hi there! How can I help you today?")]
+    
+    # Send a new message to the model
+