Update app.py

app.py

@@ -1,62 +1,147 @@
+import torch
+from PIL import Image
 import gradio as gr
-from huggingface_hub import InferenceClient
+import spaces
+from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
+import os
+from threading import Thread
 
+
+HF_TOKEN = os.environ.get("HF_TOKEN", None)
+MODEL_ID = "Qwen/Qwen2-7B-Instruct"
+MODELS = os.environ.get("MODELS")
+MODEL_NAME = MODELS.split("/")[-1]
+
+TITLE = "<h1><center>Qwen2-Vietnamese</center></h1>"
+
+DESCRIPTION = f"""
+<h3>MODEL: <a href="https://hf.co/{MODELS}">{MODEL_NAME}</a></h3>
+<center>
+<p>Qwen2 is the large language model built by Alibaba Cloud.
+<br>
+Feel free to test without log.
+</p>
+</center>
 """
-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
+
+CSS = """
+.duplicate-button {
+    margin: auto !important;
+    color: white !important;
+    background: black !important;
+    border-radius: 100vh !important;
+}
+h3 {
+    text-align: center;
+}
 """
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
+model = AutoModelForCausalLM.from_pretrained(
+          MODELS,
+          torch_dtype=torch.float16,
+          device_map="auto",
+        )
+tokenizer = AutoTokenizer.from_pretrained(MODELS)
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+@spaces.GPU
+def stream_chat(message: str, history: list, temperature: float, max_new_tokens: int, top_p: float, top_k: int, penalty: float):
+    print(f'message is - {message}')
+    print(f'history is - {history}')
+    conversation = []
+    for prompt, answer in history:
+        conversation.extend([{"role": "user", "content": prompt}, {"role": "assistant", "content": answer}])
+    conversation.append({"role": "user", "content": 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]})
+    print(f"Conversation is -\n{conversation}")
+    
+    input_ids = tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True)
+    inputs = tokenizer(input_ids, return_tensors="pt").to(0)
+    
+    streamer = TextIteratorStreamer(tokenizer, timeout=10., skip_prompt=True, skip_special_tokens=True)
 
-    messages.append({"role": "user", "content": message})
+    generate_kwargs = dict(
+        inputs, 
+        streamer=streamer,
+        top_k=top_k,
+        top_p=top_p,
+        repetition_penalty=penalty,
+        max_new_tokens=max_new_tokens, 
+        do_sample=True, 
+        temperature=temperature,
+        eos_token_id = [151645, 151643],
+    )
+    
+    thread = Thread(target=model.generate, kwargs=generate_kwargs)
+    thread.start()
 
-    response = ""
+    buffer = ""
+    for new_text in streamer:
+        buffer += new_text
+        yield buffer
 
-    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)",
-        ),
-    ],
-)
+chatbot = gr.Chatbot(height=450)
+
+with gr.Blocks(css=CSS) as demo:
+    gr.HTML(TITLE)
+    gr.HTML(DESCRIPTION)
+    gr.DuplicateButton(value="Duplicate Space for private use", elem_classes="duplicate-button")
+    gr.ChatInterface(
+        fn=stream_chat,
+        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=0.8,
+                label="Temperature",
+                render=False,
+            ),
+            gr.Slider(
+                minimum=128,
+                maximum=4096,
+                step=1,
+                value=1024,
+                label="Max new tokens",
+                render=False,
+            ),
+            gr.Slider(
+                minimum=0.0,
+                maximum=1.0,
+                step=0.1,
+                value=0.8,
+                label="top_p",
+                render=False,
+            ),
+            gr.Slider(
+                minimum=1,
+                maximum=20,
+                step=1,
+                value=20,
+                label="top_k",
+                render=False,
+            ),
+            gr.Slider(
+                minimum=0.0,
+                maximum=2.0,
+                step=0.1,
+                value=1.0,
+                label="Repetition penalty",
+                render=False,
+            ),
+        ],
+        examples=[
+            ["Viết một lá thư chúc mừng sinh nhật gửi bạn Thục Linh."],
+            ["Trường Sa và Hoàng Sa là của nước nào?"],
+            ["Giới thiệu về tỉ phú Elon Musk"],
+            ["Viết code một trang cá nhân đơn giản bằng html."],
+        ],
+        cache_examples=False,
+    )
 
 
 if __name__ == "__main__":