using paddlepaddle + qwen0.5 to demo

Signed-off-by: Zhang Jun <[email protected]>

app.py

@@ -1,63 +1,36 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from paddlenlp.transformers import AutoTokenizer, AutoModelForCausalLM
 
-"""
-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")
+tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen2-0.5B")
+model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2-0.5B", dtype="float32")
 
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+def inference(input_text):
+    print(input_text)
+    print(type(input_text))
+    input_features = tokenizer(input_text, return_tensors="pd")
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+    outputs = model.generate(**input_features, max_new_tokens=128)#max_length=128)
+    output_text = tokenizer.batch_decode(outputs[0], skip_special_tokens=True)[0]
 
-    messages.append({"role": "user", "content": message})
+    return output_text
 
-    response = ""
 
-    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
+title = 'PaddlePaddle Meets LLM'
+description = 'What is special: underlying execution is using PaddlePaddle and PaddleNLP!'
+article = "<p style='text-align: center'> PaddleNLP <a href='https://github.com/PaddlePaddle/PaddleNLP'>Github Repo</a></p>"
 
-        response += token
-        yield response
+examples = ['请自我介绍一下。', '今天吃什么好呢？']
 
-
-"""
-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)",
-        ),
-    ],
-)
+demo = gr.Interface(
+    inference,
+    inputs="text",
+    outputs="text",
+    title=title,
+    description=description,
+    article=article,
+    examples=examples,
+    )
 
 
 if __name__ == "__main__":

requirements.txt

@@ -1 +1,2 @@
-huggingface_hub==0.25.2
+paddlepaddle
+paddlenlp==3.0.0b4