stree_datasets/oc1results.py

import os
from datetime import datetime
from experimentation.Sets import Datasets
from experimentation.Database import MySQL
from statistics import mean, stdev


def store_result(database, dataset, accuracy, time_spent):
    attributes = [
        "date",
        "time",
        "type",
        "accuracy",
        "accuracy_std",
        "dataset",
        "classifier",
        "norm",
        "stand",
        "time_spent",
        "time_spent_std",
        "parameters",
    ]
    command_insert = (
        "insert into results ("
        + ",".join(attributes)
        + ") values("
        + ("%s," * len(attributes))[:-1]
        + ")"
    )
    now = datetime.now()
    date = now.strftime("%Y-%m-%d")
    time = now.strftime("%H:%M:%S")
    values = (
        date,
        time,
        "crossval",
        mean(accuracy),
        stdev(accuracy),
        dataset,
        "oc1",
        True,
        False,
        mean(time_spent),
        stdev(time_spent),
        str({"random_state": 1}),
    )
    cursor = database.cursor()
    cursor.execute(command_insert, values)
    database.commit()


def get_result(name):
    accuracy = []
    time_spent = []
    for seed in [57, 31, 1714, 17, 23, 79, 83, 97, 7, 1]:
        filename = os.path.join("data", "oc1output", f"{name}_{seed}.txt")
        result_file = open(filename, "r")
        lines = result_file.readlines()
        result_file.close()
        data = lines[-7:-2]
        for line in data:
            acc, time_s = line.split("***")[2:4]
            accuracy.append(float(acc))
            time_spent.append(float(time_s))
    return accuracy, time_spent


dbh = MySQL()
database = dbh.get_connection()
dt = Datasets(False, False, "tanveer")
for dataset in dt:
    print(f"Processing {dataset[0]:30s}", end=" ")
    accuracy, time_spent = get_result(dataset[0])
    accuracy_mean, accuracy_std = mean(accuracy), stdev(accuracy)
    time_mean, time_std = mean(time_spent), stdev(time_spent)
    print(
        f"accuracy=[{accuracy_mean:05.3f}\u00B1{accuracy_std:05.3f}] "
        f"time=[{time_mean:05.3f}\u00B1{time_std:05.3f}] "
        f"elements=[{len(accuracy)} {len(time_spent)}]"
    )
    store_result(database, dataset[0], accuracy, time_spent)
dbh.close()