132 lines
5.6 KiB
Python
132 lines
5.6 KiB
Python
import pandas as pd
|
|
import glob
|
|
import re
|
|
import os
|
|
import matplotlib.pyplot as plt
|
|
from matplotlib.ticker import AutoMinorLocator
|
|
import numpy as np
|
|
from scipy.interpolate import interp1d, CubicSpline
|
|
from scipy.optimize import curve_fit
|
|
from numpy.polynomial import Polynomial
|
|
|
|
|
|
# folder_path = 'results'
|
|
|
|
# # Список размеров (500, 1000, 2000, 5000, 10000)
|
|
# sizes = ['500', '1000', '2000', '5000', '10000']
|
|
|
|
# for size in sizes:
|
|
# files = glob.glob(os.path.join(folder_path, f'timedata_{size}_epochs_*.csv'))
|
|
|
|
# if not files:
|
|
# continue
|
|
|
|
# # Читаем файлы
|
|
# dfs = [pd.read_csv(f) for f in files]
|
|
|
|
# # Определяем, какие колонки текстовые (не числовые)
|
|
# # Предполагаем, что во всех файлах они одинаковые
|
|
# text_cols = dfs[0].select_dtypes(exclude=['number']).columns.tolist()
|
|
|
|
# # Объединяем и считаем среднее
|
|
# # Группируем по текстовым колонкам, чтобы они остались в результате
|
|
# if text_cols:
|
|
# combined = pd.concat(dfs)
|
|
# mean_df = combined.groupby(text_cols).mean().reset_index()
|
|
# else:
|
|
# # Если текста нет, просто среднее по строкам
|
|
# mean_df = pd.concat(dfs).groupby(level=0).mean()
|
|
|
|
|
|
# output_name = f'average_timedata_{size}.csv'
|
|
# mean_df.to_csv(os.path.join(folder_path, output_name), index=False)
|
|
# print(f"Файл {output_name} успешно создан")
|
|
|
|
|
|
df500 = pd.read_csv("results/average_timedata_500.csv")
|
|
df1000 = pd.read_csv("results/average_timedata_1000.csv")
|
|
df2000 = pd.read_csv("results/average_timedata_2000.csv")
|
|
df5000 = pd.read_csv("results/average_timedata_5000.csv")
|
|
df10000 = pd.read_csv("results/average_timedata_10000.csv")
|
|
|
|
|
|
def select_data_list(ax):
|
|
dfs = [df500, df1000, df2000, df5000, df10000]
|
|
Nvals = [500, 1000, 2000, 5000, 10000]
|
|
# delete, find, insert
|
|
# список:
|
|
valsSort = [list(arr[(arr['Структура'] == "linklist") & (arr['Режим'] == "sorted")]["Время (сек)"]) for arr in dfs]
|
|
valsShuff = [list(arr[(arr['Структура'] == "linklist") & (arr['Режим'] == "shuffled")]["Время (сек)"]) for arr in dfs]
|
|
# 0 - sorted 1 - shuffled
|
|
# delete
|
|
ax[0].plot(Nvals, [row[0] for row in valsSort], label="delete", color='red')
|
|
ax[1].plot(Nvals, [row[0] for row in valsShuff], color='red')
|
|
# find
|
|
ax[0].plot(Nvals, [row[1] for row in valsSort], label="find", color='blue')
|
|
ax[1].plot(Nvals, [row[1] for row in valsShuff], color='blue')
|
|
# insert
|
|
ax[0].plot(Nvals, [row[2] for row in valsSort], label="insert", color='green')
|
|
ax[1].plot(Nvals, [row[2] for row in valsShuff], color='green')
|
|
|
|
|
|
def select_data_hasht(ax):
|
|
dfs = [df500, df1000, df2000, df5000, df10000]
|
|
Nvals = [500, 1000, 2000, 5000, 10000]
|
|
# delete, find, insert
|
|
# список:
|
|
valsSort = [list(arr[(arr['Структура'] == "hashtable") & (arr['Режим'] == "sorted")]["Время (сек)"]) for arr in dfs]
|
|
valsShuff = [list(arr[(arr['Структура'] == "hashtable") & (arr['Режим'] == "shuffled")]["Время (сек)"]) for arr in dfs]
|
|
# 0 - sorted 1 - shuffled
|
|
# delete
|
|
ax[0].plot(Nvals, [row[0] for row in valsSort], label="delete", color='red')
|
|
ax[1].plot(Nvals, [row[0] for row in valsShuff], color='red')
|
|
# find
|
|
ax[0].plot(Nvals, [row[1] for row in valsSort], label="find", color='blue')
|
|
ax[1].plot(Nvals, [row[1] for row in valsShuff], color='blue')
|
|
# insert
|
|
ax[0].plot(Nvals, [row[2] for row in valsSort], label="insert", color='green')
|
|
ax[1].plot(Nvals, [row[2] for row in valsShuff], color='green')
|
|
|
|
|
|
def select_data_tree(ax):
|
|
dfs = [df500, df1000, df2000, df5000, df10000]
|
|
Nvals = [500, 1000, 2000, 5000, 10000]
|
|
# delete, find, insert
|
|
# список:
|
|
valsSort = [list(arr[(arr['Структура'] == "bintree") & (arr['Режим'] == "sorted")]["Время (сек)"]) for arr in dfs]
|
|
valsShuff = [list(arr[(arr['Структура'] == "bintree") & (arr['Режим'] == "shuffled")]["Время (сек)"]) for arr in dfs]
|
|
# 0 - sorted 1 - shuffled
|
|
# delete
|
|
ax[0].plot(Nvals, [row[0] for row in valsSort], label="delete", color='red')
|
|
ax[1].plot(Nvals, [row[0] for row in valsShuff], color='red')
|
|
# find
|
|
ax[0].plot(Nvals, [row[1] for row in valsSort], label="find", color='blue')
|
|
ax[1].plot(Nvals, [row[1] for row in valsShuff], color='blue')
|
|
# insert
|
|
ax[0].plot(Nvals, [row[2] for row in valsSort], label="insert", color='green')
|
|
ax[1].plot(Nvals, [row[2] for row in valsShuff], color='green')
|
|
|
|
# list(df500[(df500['Структура'] == "linklist") & (df500['Режим'] == "shuffled")]["Время (сек)"])
|
|
|
|
# построение графика
|
|
fig, ax = plt.subplots(figsize=(10, 5), nrows=1, ncols=2)
|
|
for i in range(2):
|
|
# select_data_list(ax)
|
|
# select_data_hasht(ax)
|
|
select_data_tree(ax)
|
|
ax[0].set_title("График сложностей для дерева (sort)")
|
|
ax[1].set_title("График сложностей для дерева (shuff)")
|
|
ax[i].set_xlabel("N")
|
|
ax[i].set_ylabel("сек * ")
|
|
ax[i].grid(which="major", linewidth=1.5)
|
|
ax[i].grid(which="minor", color="gray", linewidth=0.5)
|
|
ax[i].xaxis.set_minor_locator(AutoMinorLocator())
|
|
ax[i].yaxis.set_minor_locator(AutoMinorLocator())
|
|
ax[i].legend()
|
|
ax[i].set_ylim(0, 0.1)
|
|
plt.savefig('graphics\Tree1.png', dpi=200)
|
|
plt.savefig('graphics\Tre1.eps', dpi=200)
|
|
plt.show()
|
|
|
|
|