Update space

app.py

@@ -1,64 +1,73 @@
-import gradio as gr
-from huggingface_hub import InferenceClient
+import gradio as gr 
+from transformers import pipeline
 
-"""
-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")
+from os import getenv
+from huggingface_hub import login 
+from history import get_history, update_history
 
+# Login to Hugging Face
+login(getenv("Token"))
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+pipe = pipeline("text-generation", model="TheBloke/claude2-alpaca-7B-GGUF",trust_remote_code=True) # trust here
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+def talk(prompt,history=[]):  
+ 
+  #If no history, get history from database
+  group_name=getenv("DEFAULT_GROUP_NAME")
+  if not len(history):
+    old_history=get_history(group_name)
+    history=old_history
 
-    messages.append({"role": "user", "content": message})
+  print('the history',history)
 
-    response = ""
+  # Transform to array of tuples
+  transformed_data = []
+  user_message = None
 
-    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
+  for item in history:
+    if item['role'] == 'user':
+        user_message = item['content']  # Save user message temporarily
+    elif item['role'] == 'assistant' and user_message:
+        transformed_data.append((user_message, item['content']))
+        user_message = None  # Reset after pairing
 
-        response += token
-        yield response
+  # Output result
+  print("transformed_data",transformed_data)
 
+  response=''
+
+  try: 
+    for partial_response in pipe(prompt, history=transformed_data, stream=True, max_new_tokens=128000):
+        response = partial_response
+        yield partial_response
+  except Exception as e:
+    print("Stream queue was empty. No response generated.",str(e))
+    yield "Error: No response generated." 
+
+  #update the history in database
+  if response:
+    transformed_data.append((prompt,response))
+    update_history(group_name,transformed_data)
 
-"""
-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)",
-        ),
-    ],
-)
 
+def initialize():
+  messages=[]
+  history=get_history(getenv("DEFAULT_GROUP_NAME"))
+  for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
+  return messages
+ 
+
+demo = gr.ChatInterface(
+  fn=talk, 
+  type="messages",
+  chatbot=gr.Chatbot(value=initialize(),type="messages"), 
+  examples=[["hi",[]]],
+  cache_examples=False,
+  title="AI Interviewer 5th Iteration"
+  )
 
-if __name__ == "__main__":
-    demo.launch()
+demo.launch(share=True,ssr_mode=False)

requirements.txt

@@ -1 +1,7 @@
-huggingface_hub==0.25.2
+huggingface_hub==0.25.2
+torch
+transformers 
+accelerate>=0.26.0
+pymongo 
+
+gradio==4.44.0