Update app.py

app.py

@@ -1,19 +1,45 @@
 import gradio as gr
-from llama_cpp import Llama
 import os
-import llama_cpp_python as llama
-
-
+import requests
+import json
 
 sbc_host_url = os.environ['URL']
 
-# Create a llama client
-client = llama.Client(host="sbc_host_url", port=3344)
+def get_completion(prompt:str, messages:str = '', n_predict=128):
+    prompt_templated = f'{messages}\n ### HUMAN:\n{prompt} \n ### ASSISTANT:'
+    headers = {
+        "Content-Type": "application/json"
+    }
+    data = {
+        "prompt": prompt_templated,
+        "n_predict": n_predict,
+        "stop": ["### HUMAN:", "### ASSISTANT:", "HUMAN"],
+        "stream": "True"
+    }
+    
+    response = requests.post(sbc_host_url, headers=headers, data=json.dumps(data))
+    
+    if response.status_code == 200:
+        return response.json()['content']
+    else:
+        response.raise_for_status()
 
 
 def chatty(prompt, messages):
-    response = client.call("generate_text", prompt=prompt)
-    return response 
+    print(prompt)
+    print(f'messages: {messages}')
+    past_messages = ''
+    if len(messages) > 0:
+        for idx, message in enumerate(messages):
+            print(f'idx: {idx}, message: {message}')
+            past_messages += f'\n### HUMAN: {message[0]}'
+            past_messages += f'\n### ASSISTANT: {message[1]}'
+                
+                
+        # past_messages = messages[0][0]
+    print(f'past_messages: {past_messages}')
+    messages = get_completion(prompt, past_messages)
+    return messages.split('### ASSISTANT:')[-1]
 
 
 demo = gr.ChatInterface(