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 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, ModelTraining, fields, MalteseTraining ) 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, read_configuration 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], hide_columns=[c.name for c in fields(AutoEvalColumn) if c.hidden], filter_columns=[ ColumnFilter(AutoEvalColumn.model_training.name, type="checkboxgroup", label="Model types"), ColumnFilter(AutoEvalColumn.maltese_training.name, type="checkboxgroup", label="Maltese training"), ColumnFilter( AutoEvalColumn.language_count.name, type="slider", min=1, max=1000, label="Number of languages during training", ), ColumnFilter( AutoEvalColumn.params.name, type="slider", min=0.01, max=150, label="Select the number of parameters (B)", ), ColumnFilter(AutoEvalColumn.prompt_version.name, type="checkboxgroup", label="Prompt Version"), ColumnFilter(AutoEvalColumn.n_shot.name, type="slider", min=0, max=100, label="Number of Shots"), 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.HTML(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(): files = gr.File( label="Files (Configuration File & Prediction Outputs)", file_count="directory", type="filepath", ) with gr.Row(equal_height=True): with gr.Column(): model_name = gr.Textbox( label="Model name", info="Read automatically from the results file.", interactive=False, ) version = gr.Textbox( label="Prompt Version", info="Read automatically from the results file.", interactive=False, ) n_shots = gr.Number( label="Number of Shots", info="Read automatically from the results file.", interactive=False, ) with gr.Column(): model_training = gr.Dropdown( choices=[t.to_str(": ") for t in ModelTraining if t != ModelTraining.NK], label="Model Training", info="How to model is trained.", multiselect=False, value=None, interactive=True, ) maltese_training = gr.Dropdown( choices=[t.to_str(": ") for t in MalteseTraining if t != ModelTraining.NK], label="Maltese Training", info="The last stage of training in which Maltese was included.", multiselect=False, value=None, interactive=True, ) language_count = gr.Number( label="Number of languages", info="Include languages for all training stages. Set to 0 if unknown.", minimum=0, interactive=True, ) submit_button = gr.Button("Submit Eval") submission_result = gr.Markdown() configuration = gr.State() file_paths = gr.State() files.change(read_configuration, files, [configuration, file_paths, model_name, version, n_shots, submission_result]) submit_button.click( add_new_eval, [ model_training, maltese_training, language_count, configuration, file_paths ], 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, ) scheduler = BackgroundScheduler() scheduler.add_job(restart_space, "interval", seconds=1800) scheduler.start() demo.queue(default_concurrency_limit=40).launch()