import gradio as gr from gradio_leaderboard import Leaderboard, ColumnFilter, SelectColumns import pandas as pd from apscheduler.schedulers.background import BackgroundScheduler from huggingface_hub import snapshot_download from fastapi import FastAPI from src.api_submit_results import router as submission_router from src.about import ( CITATION_BUTTON_LABEL, CITATION_BUTTON_TEXT, EVALUATION_QUEUE_TEXT, INTRODUCTION_TEXT, LLM_BENCHMARKS_TEXT, TITLE, ) from src.display.css_html_js import custom_css from src.display.utils import ( BENCHMARK_COLS, COLS, EVAL_COLS, EVAL_TYPES, AutoEvalColumn, ModelType, fields, WeightType, Precision ) from src.envs import API, EVAL_REQUESTS_PATH, EVAL_RESULTS_PATH, QUEUE_REPO, REPO_ID, RESULTS_REPO, TOKEN from src.populate import get_evaluation_queue_df, get_leaderboard_df from src.submission.submit import add_new_eval, add_manual_results def restart_space(): API.restart_space(repo_id=REPO_ID) ### Space initialisation try: print(EVAL_REQUESTS_PATH) snapshot_download( repo_id=QUEUE_REPO, local_dir=EVAL_REQUESTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30, token=TOKEN ) except Exception: restart_space() try: print(EVAL_RESULTS_PATH) snapshot_download( repo_id=RESULTS_REPO, local_dir=EVAL_RESULTS_PATH, repo_type="dataset", tqdm_class=None, etag_timeout=30, token=TOKEN ) except Exception: restart_space() LEADERBOARD_DF = get_leaderboard_df(EVAL_RESULTS_PATH, EVAL_REQUESTS_PATH, COLS, BENCHMARK_COLS) ( finished_eval_queue_df, running_eval_queue_df, pending_eval_queue_df, ) = get_evaluation_queue_df(EVAL_REQUESTS_PATH, EVAL_COLS) def init_leaderboard(dataframe): if dataframe is None or dataframe.empty: raise ValueError("Leaderboard DataFrame is empty or None.") return Leaderboard( value=dataframe, datatype=[c.type for c in fields(AutoEvalColumn)], select_columns=SelectColumns( default_selection=[c.name for c in fields(AutoEvalColumn) if c.displayed_by_default], cant_deselect=[c.name for c in fields(AutoEvalColumn) if c.never_hidden], label="Select Columns to Display:", ), search_columns=[AutoEvalColumn.model.name, AutoEvalColumn.license.name], hide_columns=[c.name for c in fields(AutoEvalColumn) if c.hidden], filter_columns=[ ColumnFilter(AutoEvalColumn.model_type.name, type="checkboxgroup", label="Model types"), ColumnFilter(AutoEvalColumn.precision.name, type="checkboxgroup", label="Precision"), ColumnFilter( AutoEvalColumn.params.name, type="slider", min=0.01, max=150, label="Select the number of parameters (B)", ), ColumnFilter( AutoEvalColumn.still_on_hub.name, type="boolean", label="Deleted/incomplete", default=True ), ], bool_checkboxgroup_label="Hide models", interactive=False, ) demo = gr.Blocks(css=custom_css) with demo: gr.HTML(TITLE) gr.Markdown(INTRODUCTION_TEXT, elem_classes="markdown-text") with gr.Tabs(elem_classes="tab-buttons") as tabs: with gr.TabItem("🏅 LLM Benchmark", elem_id="llm-benchmark-tab-table", id=0): leaderboard = init_leaderboard(LEADERBOARD_DF) with gr.TabItem("📝 About", elem_id="llm-benchmark-tab-table", id=2): gr.Markdown(LLM_BENCHMARKS_TEXT, elem_classes="markdown-text") with gr.TabItem("🚀 Submit here! ", elem_id="llm-benchmark-tab-table", id=3): with gr.Column(): with gr.Row(): gr.Markdown(EVALUATION_QUEUE_TEXT, elem_classes="markdown-text") with gr.Column(): with gr.Accordion( f"✅ Finished Evaluations ({len(finished_eval_queue_df)})", open=False, ): with gr.Row(): finished_eval_table = gr.components.Dataframe( value=finished_eval_queue_df, headers=EVAL_COLS, datatype=EVAL_TYPES, row_count=5, ) with gr.Accordion( f"🔄 Running Evaluation Queue ({len(running_eval_queue_df)})", open=False, ): with gr.Row(): running_eval_table = gr.components.Dataframe( value=running_eval_queue_df, headers=EVAL_COLS, datatype=EVAL_TYPES, row_count=5, ) with gr.Accordion( f"⏳ Pending Evaluation Queue ({len(pending_eval_queue_df)})", open=False, ): with gr.Row(): pending_eval_table = gr.components.Dataframe( value=pending_eval_queue_df, headers=EVAL_COLS, datatype=EVAL_TYPES, row_count=5, ) with gr.Row(): gr.Markdown("# ✉️✨ Submit your model here!", elem_classes="markdown-text") with gr.Row(): with gr.Column(): model_name_textbox = gr.Textbox(label="Model name") revision_name_textbox = gr.Textbox(label="Revision commit", placeholder="main") model_type = gr.Dropdown( choices=[t.to_str(" : ") for t in ModelType if t != ModelType.Unknown], label="Model type", multiselect=False, value=None, interactive=True, ) with gr.Column(): precision = gr.Dropdown( choices=[i.value.name for i in Precision if i != Precision.Unknown], label="Precision", multiselect=False, value="float16", interactive=True, ) weight_type = gr.Dropdown( choices=[i.value.name for i in WeightType], label="Weights type", multiselect=False, value="Original", interactive=True, ) base_model_name_textbox = gr.Textbox(label="Base model (for delta or adapter weights)") submit_button = gr.Button("Submit Eval") submission_result = gr.Markdown() submit_button.click( add_new_eval, [ model_name_textbox, base_model_name_textbox, revision_name_textbox, precision, weight_type, model_type, ], submission_result, ) # ---------------------------------------------------- # Manual metrics submission form # ---------------------------------------------------- gr.Markdown("## 📝 Submit metrics manually (advanced)") with gr.Row(): with gr.Column(): model_name_metrics = gr.Textbox(label="Model name", placeholder="org/model") revision_metrics = gr.Textbox(label="Revision commit", placeholder="main", value="main") bleu_input = gr.Number(label="BLEU", value=0.5) pass1_input = gr.Number(label="Pass@1", value=0.5, minimum=0.0, maximum=1.0) pass5_input = gr.Number(label="Pass@5", value=0.5, minimum=0.0, maximum=1.0) pass10_input = gr.Number(label="Pass@10", value=0.5, minimum=0.0, maximum=1.0) with gr.Column(): # Subjective metrics sliders (0-5) readability_slider = gr.Slider(0, 5, step=1, value=3, label="Readability") relevance_slider = gr.Slider(0, 5, step=1, value=3, label="Relevance") explanation_slider = gr.Slider(0, 5, step=1, value=3, label="Explanation clarity") problem_slider = gr.Slider(0, 5, step=1, value=3, label="Problem identification") actionability_slider = gr.Slider(0, 5, step=1, value=3, label="Actionability") completeness_slider = gr.Slider(0, 5, step=1, value=3, label="Completeness") specificity_slider = gr.Slider(0, 5, step=1, value=3, label="Specificity") contextual_slider = gr.Slider(0, 5, step=1, value=3, label="Contextual adequacy") consistency_slider = gr.Slider(0, 5, step=1, value=3, label="Consistency") brevity_slider = gr.Slider(0, 5, step=1, value=3, label="Brevity") submit_metrics_button = gr.Button("Submit Metrics") metrics_submission_result = gr.Markdown() submit_metrics_button.click( add_manual_results, [ model_name_metrics, revision_metrics, bleu_input, readability_slider, relevance_slider, explanation_slider, problem_slider, actionability_slider, completeness_slider, specificity_slider, contextual_slider, consistency_slider, brevity_slider, pass1_input, pass5_input, pass10_input, ], metrics_submission_result, ) with gr.Row(): with gr.Accordion("📙 Citation", open=False): citation_button = gr.Textbox( value=CITATION_BUTTON_TEXT, label=CITATION_BUTTON_LABEL, lines=20, elem_id="citation-button", show_copy_button=True, ) # ------------------------------ # Start background scheduler # ------------------------------ scheduler = BackgroundScheduler() scheduler.add_job(restart_space, "interval", seconds=1800) scheduler.start() # ------------------------------ # Mount Gradio UI into FastAPI application # ------------------------------ # Removed direct .launch(); Gradio UI will be served via the mounted FastAPI `app`. # ------------------ FastAPI mounting ------------------ backend = FastAPI() backend.include_router(submission_router) # Enable queuing (same limit as before) demo = demo.queue(default_concurrency_limit=40) # Expose `app` for the HF Spaces runtime app = gr.mount_gradio_app(backend, demo, path="/")