import json import os import shutil import gradio as gr from huggingface_hub import Repository from text_generation import Client from share_btn import community_icon_html, loading_icon_html, share_js, share_btn_css HF_TOKEN = os.environ.get("HF_TOKEN", None) API_URL = os.environ.get("API_URL") theme = gr.themes.Monochrome( primary_hue="indigo", secondary_hue="blue", neutral_hue="slate", radius_size=gr.themes.sizes.radius_sm, font=[gr.themes.GoogleFont("Open Sans"), "ui-sans-serif", "system-ui", "sans-serif"], ) client = Client( API_URL, #headers={"Authorization": f"Bearer {HF_TOKEN}"}, ) def generate(prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0): temperature = float(temperature) if temperature < 1e-2: temperature = 1e-2 top_p = float(top_p) generate_kwargs = dict( temperature=temperature, max_new_tokens=max_new_tokens, top_p=top_p, repetition_penalty=repetition_penalty, do_sample=True, truncate=999, seed=42, stop_sequences=[""], ) stream = client.generate_stream( prompt, **generate_kwargs, ) output = "" for response in stream: output += response.token.text yield output return output examples = [ "def hello_world():", ] def process_example(args): for x in generate(args): pass return x css = ".generating {visibility: hidden}" + share_btn_css with gr.Blocks(theme=theme, analytics_enabled=False, css=css) as demo: with gr.Column(): gr.Markdown( """ # BigCode - Playground """ ) with gr.Row(): with gr.Column(scale=3): instruction = gr.Textbox(placeholder="Enter your prompt here", label="Prompt", elem_id="q-input") with gr.Box(): gr.Markdown("**Response**") output = gr.Code(elem_id="q-output") submit = gr.Button("Generate", variant="primary") gr.Examples( examples=examples, inputs=[instruction], cache_examples=False, fn=process_example, outputs=[output], ) with gr.Column(scale=1): temperature = gr.Slider( label="Temperature", value=0.9, minimum=0.0, maximum=2.0, step=0.1, interactive=True, info="Higher values produce more diverse outputs", ) max_new_tokens = gr.Slider( label="Max new tokens", value=256, minimum=0, maximum=512, step=4, interactive=True, info="The maximum numbers of new tokens", ) top_p = gr.Slider( label="Top-p (nucleus sampling)", value=0.90, minimum=0.0, maximum=1, step=0.05, interactive=True, info="Higher values sample more low-probability tokens", ) repetition_penalty = gr.Slider( label="Repetition penalty", value=1.2, minimum=1.0, maximum=2.0, step=0.05, interactive=True, info="Penalize repeated tokens", ) submit.click(generate, inputs=[instruction, temperature, max_new_tokens, top_p, repetition_penalty], outputs=[output]) instruction.submit(generate, inputs=[instruction, temperature, max_new_tokens, top_p, repetition_penalty], outputs=[output]) demo.queue(concurrency_count=16).launch(debug=True)