Changed the app.py.

.gitignore

@@ -0,0 +1 @@
+.ipynb_checkpoints/

app.py

@@ -1,76 +1,88 @@
-import gradio as gr
 from huggingface_hub import InferenceClient
 import os
 
 HF_TOKEN = os.environ.get("HF_TOKEN", None)
 
-"""
-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("meta-llama/Meta-Llama-3-8B-Instruct")
-
+client = InferenceClient(
+    "meta-llama/Meta-Llama-3-8B-Instruct"
+)
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
+punctuation_marks = [".", "!", "?"]
+def generate(
+    prompt, history, temperature=0.2, max_new_tokens=256, top_p=0.8, repetition_penalty=1.0,
 ):
-    messages = [{"role": "system", "content": system_message}]
+    temperature = float(temperature)
+    if temperature < 1e-2:
+        temperature = 1e-2
+    top_p = float(top_p)
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+    generate_kwargs = dict(
+        temperature=temperature,
+        max_new_tokens=max_new_tokens,
+        top_p=top_p,
+        repetition_penalty=repetition_penalty,
+        do_sample=True,
+        seed=42,
+    )
 
-    messages.append({"role": "user", "content": message})
 
-    response = ""
+    stream = client.text_generation(prompt, **generate_kwargs, stream=True, details=True, return_full_text=False)
+    output = ""
 
-    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 response in stream:
+        output += response.token.text
+    while output and output[-1] not in punctuation_marks:
+      output = output[:-1]
+        # yield output
+    return output
 
-        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(
-    fn=respond,
-    chatbot=gr.Chatbot(show_label=False, show_share_button=False,
-                       show_copy_button=True, likeable=True, layout="panel"),
-    title="""Have a chat with LLama3 8B""",
-    additional_inputs=[
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens",interactive=True,
-        info="The maximum numbers of new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.1, step=0.1, label="Temperature",interactive=True,
-        info="Higher values produce more diverse outputs"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
+additional_inputs=[
+    gr.Slider(
+        label="Temperature",
+        value=0.2,
+        minimum=0.0,
+        maximum=1.0,
+        step=0.05,
+        interactive=True,
+        info="Higher values produce more diverse outputs",
+    ),
+    gr.Slider(
+        label="Max new tokens",
+        value=256,
+        minimum=0,
+        maximum=1048,
+        step=64,
         interactive=True,
-        info="Higher values sample more low-probability tokens"),
-        gr.Textbox(value="You are a truthful and friendly Assistant.", label="System message"),
-    ],
+        info="The maximum numbers of new tokens",
+    ),
+    gr.Slider(
+        label="Top-p (nucleus sampling)",
+        value=0.80,
+        minimum=0.0,
+        maximum=1,
+        step=0.05,
+        interactive=True,
+        info="Higher values sample more low-probability tokens",
+    ),
+    gr.Slider(
+        label="Repetition penalty",
+        value=1.0,
+        minimum=1.0,
+        maximum=2.0,
+        step=0.05,
+        interactive=True,
+        info="Penalize repeated tokens",
+    )
+]
+
+
+gr.ChatInterface(
+    fn=generate,
+    chatbot=gr.Chatbot(show_label=False, show_share_button=False, show_copy_button=True, likeable=True, layout="panel"),
+    additional_inputs=additional_inputs,
     examples=[
         ["Can you explain briefly to me what is the Python programming language?"],
         ["Write a 100-word article on 'Benefits of Open-Source in AI research'"],
     ]
-)
-
-
-if __name__ == "__main__":
-    demo.launch(show_api=False)
+).launch(show_api=False)