Change mld output file from detect_regressions.py

import plotly.graph_objects as go

from plotly.subplots import make_subplots

def main(logs_dir, output_filename, measure, days):

def main(args):

    logs_dir = args.logs_dir

    output_filename = args.output_filename

    measure = args.measure

    days = args.days

    input_path = Path(logs_dir)

    logs = [f for f in input_path.iterdir() if f.is_dir()]

    for date, jobs in logdict.items():

        for job, testcases in jobs.items():

            for testcase, value in testcases.items():

                csv_rows.append((job, testcase, formatdict[date][job][testcase], date, value))

                csv_rows.append(

                    (job, testcase, formatdict[date][job][testcase], date, value)

                )

    result = pd.DataFrame(csv_rows, columns=["job", "testcase", "format", "date", "value"])

    result = pd.DataFrame(

        csv_rows, columns=["job", "testcase", "format", "date", "value"]

    )

    result = result.pivot(

        index=["job", "testcase", "format"], columns="date", values="value"

    ).reset_index()

    # Keep only tests for which results exist in the last run

    # Keep only tests for which results exist in any of the days

    if days == -1:

        rng = result.columns[3:]  # Whole history

    else:

    values = result.iloc[:, 3:]

    date = values.columns

    formats = result['format'].dropna().unique().tolist()

    formats = result["format"].dropna().unique().tolist()

    plotdata = pd.DataFrame(0.0, index=formats, columns=dates[-(days+1):-1])

    plottext = pd.DataFrame("", index=formats, columns=dates[-(days+1):-1])

    plotdata = pd.DataFrame(0.0, index=formats, columns=dates[-(days) : -1])

    plottext = pd.DataFrame("", index=formats, columns=dates[-(days) : -1])

    all_indices = []

    for i in range(days):

        currdate = dates[-(days-i+1)] # Make robust for shorter history

        prevdate = dates[-(days-i+2)]

        currdate = dates[-(days - i)]  # Make robust for shorter history

        prevdate = dates[-(days - i + 1)]

        idx = ratio.groupby("format")[currdate].idxmax()

        all_indices.append(idx)

        all_indices.extend(idx.tolist())

        # Store worst case per format for plotting

        for f in formats:

            plotdata.loc[f, currdate] = ratio.iloc[idx[f]][currdate]

            plottext.loc[f, currdate] = f"Job: {result.iloc[idx[f]]['job']}<br>Testcase: {result.iloc[idx[f]]['testcase']} <br>Max {measure} ratio: {ratio.iloc[idx[f]][currdate]:.2f}<br>Date: {currdate}"

            plottext.loc[f, currdate] = (

                f"Job: {result.iloc[idx[f]]['job']}<br>Testcase: {result.iloc[idx[f]]['testcase']} <br>Max {measure} ratio: {ratio.iloc[idx[f]][currdate]:.2f}<br>Date: {currdate}"

            )

    fig = make_subplots(

        rows=5,

    # Save to html

    fig.write_html(output_filename)

    # Write CSV-file

    if args.csv:

        output = result.iloc[all_indices].copy()

        cols = ["job","testcase","format"]

        cols.extend(date[-days:].tolist())

        output = output.loc[:,cols]

        values = output.iloc[:, 3:]

        last_date = values.columns[-1]

        output.insert(3, "min_date", values.idxmin(axis=1))

        output.insert(4, "min_sha", output["min_date"].map(sha))

        output.insert(5, "curr_value", output[last_date])

        output.insert(6, "min_value", values.min(axis=1))

        output.insert(7, "diff", output["curr_value"] - output["min_value"])

        output.insert(8, "ratio", output["curr_value"] / output["min_value"])

        output.loc[output["min_value"] == 0, "ratio"] = (

            1  # Set ratio to 1 for denominator 0

        )

        output["min_sha"] = (

            "'" + output["min_sha"]

        )  # Add apostrophy to prevent Excel reading this as a number

        output.sort_values(

            by=["format", "ratio"], ascending=[True, False], inplace=True

        )

        output.to_csv(args.csv, sep=";", index=False)

if __name__ == "__main__":

    parser = argparse.ArgumentParser(description="logs dir")

    parser.add_argument(

        help="Number of days in history, (default: whole history)",

        default=-1,

    )

    parser.add_argument(

        "--csv",

        type=str,

        help="CSV output file",

    )

    args = parser.parse_args()

    main(args.logs_dir, args.output_filename, args.measure, args.days)

    main(args)