Update app.py

app.py

@@ -1,64 +1,139 @@
+# Ref: https://huggingface.co/spaces/ysharma/Chat_with_Meta_llama3_8b
+
 import gradio as gr
-from huggingface_hub import InferenceClient
+import os
+import spaces
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
+from threading import Thread
 
+DESCRIPTION = '''
+<div>
+<h1 style="text-align: center;">Sarashina2.2-3B Instruct</h1>
+<p>非公式Sarashina2.2-3B Instructだよ。 <a href="https://huggingface.co/sbintuitions/sarashina2.2-3b-instruct-v0.1"><b>sbintuitions/sarashina2.2-3b-instruct-v0.1</b></a>.</p>
+</div>
+'''
+
+LICENSE = """
+<p/>
+
+---
+Built with Sarashina
 """
-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")
 
+PLACEHOLDER = """
+<div style="padding: 30px; text-align: center; display: flex; flex-direction: column; align-items: center;">
+   <h1 style="font-size: 28px; margin-bottom: 2px; opacity: 0.55;">非公式 Sarashina2.2-3B Instruct Test</h1>
+   <p style="font-size: 18px; margin-bottom: 2px; opacity: 0.65;">なんでもきいてね</p>
+</div>
+"""
 
-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]})
+css = """
+h1 {
+  text-align: center;
+  display: block;
+}
 
-    messages.append({"role": "user", "content": message})
+#duplicate-button {
+  margin: auto;
+  color: white;
+  background: #1565c0;
+  border-radius: 100vh;
+}
+"""
 
-    response = ""
+# Load the tokenizer and model
+tokenizer = AutoTokenizer.from_pretrained("sbintuitions/sarashina2.2-3b-instruct-v0.1", use_fast=False)
+model = AutoModelForCausalLM.from_pretrained("sbintuitions/sarashina2.2-3b-instruct-v0.1", torch_dtype=torch.bfloat16)
+model=model.to("cuda:0")
 
-    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
+@spaces.GPU()
+def chat_llama3_8b(message: str, 
+              history: list, 
+              temperature: float, 
+              max_new_tokens: int
+             ) -> str:
+    """
+    Generate a streaming response using the llama3-8b model.
+    Args:
+        message (str): The input message.
+        history (list): The conversation history used by ChatInterface.
+        temperature (float): The temperature for generating the response.
+        max_new_tokens (int): The maximum number of new tokens to generate.
+    Returns:
+        str: The generated response.
+    """
+    conversation = []
+    for user, assistant in history:
+        conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
+    conversation.append({"role": "user", "content": message})
 
-        response += token
-        yield response
+    input_ids = tokenizer.apply_chat_template(conversation, add_generation_prompt=True,return_tensors="pt")
+    
+    streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
 
+    generate_kwargs = dict(
+        input_ids= input_ids.to(model.device),
+        streamer=streamer,
+        max_new_tokens=max_new_tokens,
+        do_sample=True,
+        temperature=temperature,
+        top_p=0.95,
+        repetition_penalty=1.1,
+    )
+    # This will enforce greedy generation (do_sample=False) when the temperature is passed 0, avoiding the crash.             
+    if temperature == 0:
+        generate_kwargs['do_sample'] = False
+        
+    t = Thread(target=model.generate, kwargs=generate_kwargs)
+    t.start()
 
-"""
-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)",
-        ),
-    ],
-)
+    outputs = []
+    for text in streamer:
+        outputs.append(text)
+        print(outputs)
+        yield "".join(outputs)
+        
 
+# Gradio block
+chatbot=gr.Chatbot(height=450, placeholder=PLACEHOLDER, label='Gradio ChatInterface')
 
+with gr.Blocks(fill_height=True, css=css) as demo:
+    
+    gr.Markdown(DESCRIPTION)
+    gr.DuplicateButton(value="Duplicate Space for private use", elem_id="duplicate-button")
+    gr.ChatInterface(
+        fn=chat_llama3_8b,
+        chatbot=chatbot,
+        fill_height=True,
+        additional_inputs_accordion=gr.Accordion(label="⚙️ Parameters", open=False, render=False),
+        additional_inputs=[
+            gr.Slider(minimum=0,
+                      maximum=1, 
+                      step=0.1,
+                      value=0.5, 
+                      label="Temperature", 
+                      render=False),
+            gr.Slider(minimum=128, 
+                      maximum=4096,
+                      step=1,
+                      value=512, 
+                      label="Max new tokens", 
+                      render=False ),
+            ],
+        examples=[
+            ['小学生にもわかるように相対性理論を教えてください。'],
+            ['宇宙の起源を知るための方法をステップ・バイ・ステップで教えてください。'],
+            ['1から100までの素数を求めるスクリプトをPythonで書いてください。'],
+            ['友達の陽葵にあげる誕生日プレゼントを考えてください。ただし、陽葵は中学生で、私は同じクラスの男性であることを考慮してください。'],
+            ['ペンギンがジャングルの王様であることを正当化するように説明してください。']
+            ],
+        cache_examples=False,
+                     )
+    
+    gr.Markdown(LICENSE)
+    
 if __name__ == "__main__":
     demo.launch()
+