import gradio as gr
from transformers import GPT2Tokenizer, GPT2Model
from transformers import pipeline, set_seed

def inference(text):
    model_path = "distilgpt2"
    generator = pipeline('text-generation', model=model_path)
    set_seed(42)

    output=[]
    lst=generator(text, max_length=20, num_return_sequences=5)
    for dic in lst:
        output.append(dic['generated_text'])

    return output

    # tokenizer = GPT2Tokenizer.from_pretrained(model_path)
    # model = GPT2Model.from_pretrained(model_path)
    # encoded_input = tokenizer(text, return_tensors='pt')
    # output = model(**encoded_input)
    # print(output)
    # return output


examples=[["Hello, I’m a language model."]]

with gr.Blocks() as demo:
    gr.Markdown(
    """
    # Text generation:distilgpt2
    Gradio Demo for distilgpt2. To use it, simply type in text, or click one of the examples to load them.
    """)
    with gr.Row():
        text_input = gr.Textbox()
        text_output = gr.Textbox()
    image_button = gr.Button("上传")
    image_button.click(inference, inputs=text_input, outputs=text_output)
    gr.Examples(examples,inputs=text_input)

demo.launch()