fork download
copy 
  1. import statsmodels.formula.api as smf
  2. import seaborn as sns
  3. import matplotlib.pyplot as plt
  4. from pandas import DataFrame
  5. import tkinter as tk
  6. import tkinter.filedialog as fd
  7. import csv
  8. import pandas as pd
  9. import collections
  10. import numpy as np
  11.  
  12.  
  13. def outlier_iqr(df, chk_column, output_column):
  14.  
  15.     # 列を抽出する
  16.     col = df.ix[:, chk_column]
  17.     print(col)
  18.     # 四分位数
  19.     # q1 = col.describe()['25%']
  20.     # q3 = col.describe()['75%']
  21.     q75, q25 = np.percentile(col, [75, 25])
  22.  
  23.     iqr = q75 - q25  # 四分位範囲
  24.     print("25パーセント点", q25)
  25.     print("75パーセント点", q75)
  26.     print("四分位範囲", iqr)
  27.  
  28.     # 外れ値の基準点
  29.     outlier_min = q25 - (iqr) * 1.5
  30.     outlier_max = q75 + (iqr) * 1.5
  31.  
  32.     print('aa', outlier_min)
  33.     print(outlier_max)
  34.  
  35.     print(df[chk_column])
  36.  
  37.     # 範囲から外れている値を除く
  38.     for i, latency in enumerate(df[chk_column]):
  39.         if latency > outlier_max or latency < outlier_min:
  40.             df.iloc[i][output_column] = 1
  41.             print('1')
  42.             print(df.iloc[i][output_column])
  43.         else:
  44.             df.iloc[i][output_column] = 0
  45.             print('0')
  46.             print(df.iloc[i][output_column])
  47.     return df
  48.  
  49.  
  50. def input_data():
  51.     root = tk.Tk()
  52.     root.withdraw()
  53.     file = fd.askopenfilename(
  54.         title="ファイルを選んでください",
  55.         filetypes=[("TEXT", "csv"), ("TEXT", "py"), ("HTML", "html")]
  56.     )
  57.  
  58.     df = pd.read_csv(file)
  59.  
  60.     return df
  61.  
  62.  
  63. df = input_data()
  64. df = outlier_iqr(df, 'latency_msec', 'outlier_index')
  65. print(df)
Runtime error #stdin #stdout #stderr 0.01s 7208KB
comments (?)
stdin
copy 
x	latency_msec	outlier_index
1	1700	0
3	2500	0
545	233	0
2	565	0
4	2.8	0
1	1458	0
2	215	0
3	25488	0
4	245	0
stdout
copy 
Standard output is empty
stderr
copy 
Traceback (most recent call last):
  File "prog.py", line 1, in <module>
ImportError: No module named statsmodels.formula.api
https://ideone.com/Fl1R9z
language:
Python (cpython 2.7.16)
created:
4 years ago
visibility:
public

Share or Embed source code

Discover > Sphere Engine API

The brand new service which powers Ideone!

Discover > IDE Widget

Widget for compiling and running the source code in a web browser!