app.py

import sys
sys.path.append("src")
from src import load_pretrained, ModelArguments

# if __name__ == "__main__":
model_args = ModelArguments(checkpoint_dir="temp")
model, tokenizer = load_pretrained(model_args)
model = model.half().cuda()
model = model.eval()

import time

def predict(input, history=None):
    if history is None:
        history = []
    response, _ = model.chat(tokenizer, input, history)
    history[-1][1] = ""
    for character in response:
        history[-1][1] += character
        time.sleep(0.05)
        yield history

with gr.Blocks() as demo:
    gr.Markdown('''## DigitalWDF - unofficial demo
    ''')
    chatbot = gr.Chatbot([], elem_id="chatbot").style(height=200)
    
    
    def user(user_message, history):
        return history + [[user_message, None]]
    
    with gr.Row():
        with gr.Column(scale=4):
            txt = gr.Textbox(show_label=False, placeholder="Enter text and press enter").style(container=False)
        with gr.Column(scale=1):
            button = gr.Button("Generate")
    txt.submit(user, [txt, chatbot], chatbot, queue=False).then(predict, [txt, chatbot], chatbot)
    button.click(user, [txt, chatbot], chatbot, queue=False).then(predict, [txt, chatbot], chatbot)
demo.queue().launch()