CompileError/WebCoderBench

1import json2import os3 4import pandas as pd5 6from src.display.formatting import has_no_nan_values, make_clickable_model7from src.display.utils import AutoEvalColumn, EvalQueueColumn8from src.leaderboard.read_evals import get_raw_eval_results9 10 11def get_leaderboard_df(results_path: str, requests_path: str, cols: list, benchmark_cols: list) -> pd.DataFrame:12    """Creates a dataframe from all the individual experiment results"""13    raw_data = get_raw_eval_results(results_path, requests_path)14    all_data_json = [v.to_dict() for v in raw_data]15 16    df = pd.DataFrame.from_records(all_data_json)17    df = df.sort_values(by=[AutoEvalColumn.average.name], ascending=False)18    df = df[cols].round(decimals=2)19 20    # filter out if any of the benchmarks have not been produced21    df = df[has_no_nan_values(df, benchmark_cols)]22    return df23 24 25def get_evaluation_queue_df(save_path: str, cols: list) -> list[pd.DataFrame]:26    """Creates the different dataframes for the evaluation queues requestes"""27    entries = [entry for entry in os.listdir(save_path) if not entry.startswith(".")]28    all_evals = []29 30    for entry in entries:31        if ".json" in entry:32            file_path = os.path.join(save_path, entry)33            with open(file_path) as fp:34                data = json.load(fp)35 36            data[EvalQueueColumn.model.name] = make_clickable_model(data["model"])37            data[EvalQueueColumn.revision.name] = data.get("revision", "main")38 39            all_evals.append(data)40        elif ".md" not in entry:41            # this is a folder42            sub_entries = [43                e for e in os.listdir(f"{save_path}/{entry}") if os.path.isfile(e) and not e.startswith(".")44            ]45            for sub_entry in sub_entries:46                file_path = os.path.join(save_path, entry, sub_entry)47                with open(file_path) as fp:48                    data = json.load(fp)49 50                data[EvalQueueColumn.model.name] = make_clickable_model(data["model"])51                data[EvalQueueColumn.revision.name] = data.get("revision", "main")52                all_evals.append(data)53 54    pending_list = [e for e in all_evals if e["status"] in ["PENDING", "RERUN"]]55    running_list = [e for e in all_evals if e["status"] == "RUNNING"]56    finished_list = [e for e in all_evals if e["status"].startswith("FINISHED") or e["status"] == "PENDING_NEW_EVAL"]57    df_pending = pd.DataFrame.from_records(pending_list, columns=cols)58    df_running = pd.DataFrame.from_records(running_list, columns=cols)59    df_finished = pd.DataFrame.from_records(finished_list, columns=cols)60    return df_finished[cols], df_running[cols], df_pending[cols]61