import gradio as gr from huggingface_hub import InferenceClient # import torch # from transformers import pipeline from prometheus_client import start_http_server, Counter, Summary from typing import Iterable from gradio.themes.base import Base from gradio.themes.utils import colors, fonts, sizes # Prometheus metrics REQUEST_COUNTER = Counter('app_requests_total', 'Total number of requests') SUCCESSFUL_REQUESTS = Counter('app_successful_requests_total', 'Total number of successful requests') FAILED_REQUESTS = Counter('app_failed_requests_total', 'Total number of failed requests') REQUEST_DURATION = Summary('app_request_duration_seconds', 'Time spent processing request') # import os # from dotenv import load_dotenv # load_dotenv() # # HF_ACCESS = os.getenv("HF_ACCESS") # Inference client setup client = InferenceClient(model="mistralai/Mistral-Small-Instruct-2409", # token=HF_ACCESS ) # pipe = pipeline("text-generation", "microsoft/Phi-3-mini-4k-instruct", torch_dtype=torch.bfloat16, device_map="auto") # Global flag to handle cancellation stop_inference = False def respond( message, history: list[tuple[str, str]], system_message="You are a friendly and playful cat. Answer all user queries clearly and engagingly", max_tokens=512, temperature=0.7, top_p=0.95, use_local_model=False, ): system_message += " You also love puns and add 'meow' at the end of every response." global stop_inference stop_inference = False # Reset cancellation flag REQUEST_COUNTER.inc() # Increment request counter request_timer = REQUEST_DURATION.time() # Start timing the request try: # Initialize history if it's None if history is None: history = [] # API-based inference messages = [{"role": "system", "content": system_message}] for val in history: if val[0]: messages.append({"role": "user", "content": val[0]}) if val[1]: messages.append({"role": "assistant", "content": val[1]}) messages.append({"role": "user", "content": message}) response = "" for message_chunk in client.chat_completion( messages, max_tokens=max_tokens, stream=False, temperature=temperature, top_p=top_p, ): if stop_inference: response = "Inference cancelled." yield history + [(message, response)] return if stop_inference: response = "Inference cancelled." break token = message_chunk.choices[0].delta.content response += token yield history + [(message, response)] # Yield history + new response SUCCESSFUL_REQUESTS.inc() # Increment successful request counter except Exception as e: FAILED_REQUESTS.inc() # Increment failed request counter yield history + [(message, f"Error: {str(e)}")] finally: request_timer.observe_duration() # Stop timing the request def cancel_inference(): global stop_inference stop_inference = True # Custom CSS for a fancy look custom_css = """ #main-container { background-color: #FFC0CB; background-image: url('file=image.ipg'); font-family: 'Arial', sans-serif; } .gradio-container { max-width: 700px; margin: 0 auto; padding: 20px; background: #FFC0CB; box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1); border-radius: 10px; } .gr-button { background-color: #4CAF50; color: white; border: none; border-radius: 5px; padding: 10px 20px; cursor: pointer; transition: background-color 0.3s ease; } .gr-button:hover { background-color: #45a049; } .gr-slider input { color: #4CAF50; } .gr-chat { font-size: 16px; } #title { text-align: center; font-size: 2em; margin-bottom: 20px; color: #333; } """ class UI_design(Base): def __init__( self, *, primary_hue: colors.Color | str = colors.emerald, secondary_hue: colors.Color | str = colors.blue, neutral_hue: colors.Color | str = colors.blue, spacing_size: sizes.Size | str = sizes.spacing_md, radius_size: sizes.Size | str = sizes.radius_md, text_size: sizes.Size | str = sizes.text_lg, font: fonts.Font | str | Iterable[fonts.Font | str] = ( fonts.GoogleFont("Quicksand"), "ui-sans-serif", "sans-serif", ), font_mono: fonts.Font | str | Iterable[fonts.Font | str] = ( fonts.GoogleFont("IBM Plex Mono"), "ui-monospace", "monospace", ), ): super().__init__( primary_hue=primary_hue, secondary_hue=secondary_hue, neutral_hue=neutral_hue, spacing_size=spacing_size, radius_size=radius_size, text_size=text_size, font=font, font_mono=font_mono, ) super().set( body_background_fill="repeating-linear-gradient(45deg, *primary_200, *primary_200 10px, *primary_50 10px, *primary_50 20px)", body_background_fill_dark="repeating-linear-gradient(45deg, *primary_800, *primary_800 10px, *primary_900 10px, *primary_900 20px)", button_primary_background_fill="linear-gradient(90deg, *primary_300, *secondary_400)", button_primary_background_fill_hover="linear-gradient(90deg, *primary_200, *secondary_300)", button_primary_text_color="white", button_primary_background_fill_dark="linear-gradient(90deg, *primary_600, *secondary_800)", slider_color="*secondary_300", slider_color_dark="*secondary_600", block_title_text_weight="600", block_border_width="3px", block_shadow="*shadow_drop_lg", button_shadow="*shadow_drop_lg", button_large_padding="32px", ) ui_design = UI_design() # Define the interface # with gr.Blocks(theme=ui_design) as demo: with gr.Blocks(css=custom_css) as demo: gr.Markdown("

😸 Meowthamatical AI Chatbot 😸

") gr.Markdown(" Welcome to the Cat & Math Chatbot! Whether you're here to sharpen your math skills or just enjoy some cat-themed fun, we're excited to make learning a little more pawsome!!") # with gr.Row(): # with gr.Column(): # with gr.Tabs() as input_tabs: # with gr.Tab("Sketch"): # input_sketchpad = gr.Sketchpad(type="pil", label="Sketch", layers=False) # # input_text = gr.Textbox(label="input your question") # # with gr.Row(): # # with gr.Column(): # # clear_btn = gr.ClearButton( # # [input_sketchpad, input_text]) # with gr.Column(): # submit_btn = gr.Button("Submit", variant="primary") with gr.Row(): system_message = gr.Textbox(value="You are a friendly and playful cat who loves help users learn math.", label="System message", interactive=True) use_local_model = gr.Checkbox(label="Use Local Model", value=False) # button_1 = gr.Button("Submit", variant="primary") with gr.Row(): max_tokens = gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens") temperature = gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature") top_p = gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)") chat_history = gr.Chatbot(label="Chat") user_input = gr.Textbox(show_label=False, placeholder="Type your message here...") cancel_button = gr.Button("Cancel Inference", variant="danger") # Adjusted to ensure history is maintained and passed correctly user_input.submit(respond, [user_input, chat_history, system_message, max_tokens, temperature, top_p, use_local_model], chat_history) # user_input.submit(respond, # [user_input, chat_history, system_message, 512, 0.8, 0.95, use_local_model], # chat_history) cancel_button.click(cancel_inference) if __name__ == "__main__": start_http_server(8000) # Expose metrics on port 8000 demo.launch(share=False) # Remove share=True because it's not supported on HF Spaces