dataframe_plotter demo

This notebook demonstrates the options available with dataframe_plotter



In [1]:

    
import numpy as np
import pandas as pd

from dataframe_plotter import dataframe_plotter
%matplotlib inline

Make a DataFrame to play with



In [2]:

    
index = pd.date_range('2016-01-01', '2017-01-01', freq='H')
df = pd.DataFrame(np.arange(len(index)), index=index, columns=['linear'])

x = np.arange(len(df))
df['sin_year'] = np.sin(1 * 2 * np.pi * x / len(df))
df['sin_month'] = np.sin(12 * 2 * np.pi * x / len(df))
df['sin_day'] = np.sin(365 * 2 * np.pi * x / len(df))
df['cos_year'] = np.cos(1 * 2 * np.pi * x / len(df))
df['cos_month'] = np.cos(12 * 2 * np.pi * x / len(df))
df['cos_day'] = np.cos(365 * 2 * np.pi * x / len(df))
df['10sin_month'] = 10 * df['sin_month'] + df['sin_day']
df['10cos_month'] = 10 * df['cos_month'] + df['cos_day']
df.head()









    Out[2]:






  
    
      
      linear
      sin_year
      sin_month
      sin_day
      cos_year
      cos_month
      cos_day
      10sin_month
      10cos_month
    
  
  
    
      2016-01-01 00:00:00
      0
      0.000000
      0.000000
      0.000000
      1.000000
      1.000000
      1.000000
      0.000000
      11.000000
    
    
      2016-01-01 01:00:00
      1
      0.000715
      0.008583
      0.258099
      1.000000
      0.999963
      0.966118
      0.343924
      10.965750
    
    
      2016-01-01 02:00:00
      2
      0.001430
      0.017164
      0.498709
      0.999999
      0.999853
      0.866769
      0.670353
      10.865296
    
    
      2016-01-01 03:00:00
      3
      0.002146
      0.025745
      0.705525
      0.999998
      0.999669
      0.708685
      0.962974
      10.705371
    
    
      2016-01-01 04:00:00
      4
      0.002861
      0.034324
      0.864532
      0.999996
      0.999411
      0.502579
      1.207768
      10.496686

No subplots



In [3]:

    
dataframe_plotter(df, title='Single Plot', ylabel='All columns', xlabel='Datetime')

Suplots=True

Mirrors the operation of df.plot(subplots=True) with the added ability to have titles for the individual subplots.



In [4]:

    
dataframe_plotter(df, subplots=True, title=['First plot', 'Seconds plot'], xlabel='Datetime',
                  x_label_rotation=30)

Specify columns for the subplots



In [5]:

    
subplots = ['linear', ['sin_month', 'cos_month']]
title = ['Linear', 'Monthly sin and cos']
dataframe_plotter(df, subplots=subplots, title=title, xlabel='Datetime', dots=False)

Secondary y axis



In [6]:

    
subplots = ['linear', 'sin_month', 'cos_month']
ylabel = ['lin', 'sin', 'cos']
secondary_y = ['', ['10sin_month', '10cos_month']]
secondary_y_label = ['', '10*sin and 10*cos with noise']
dataframe_plotter(df, subplots=subplots, ylabel=ylabel,
                  secondary_y=secondary_y, secondary_ylabel=secondary_y_label,
                  dots=False, linewidth=2, height_per_plot=4)

Average value at each time of day



In [7]:

    
dataframe_plotter(df, plot_type='average', title='Plot average value at time of day',
                  subplots=[['sin_day', 'cos_day']], xlabel='hour in day')

	linear	sin_year	sin_month	sin_day	cos_year	cos_month	cos_day	10sin_month	10cos_month
2016-01-01 00:00:00	0	0.000000	0.000000	0.000000	1.000000	1.000000	1.000000	0.000000	11.000000
2016-01-01 01:00:00	1	0.000715	0.008583	0.258099	1.000000	0.999963	0.966118	0.343924	10.965750
2016-01-01 02:00:00	2	0.001430	0.017164	0.498709	0.999999	0.999853	0.866769	0.670353	10.865296
2016-01-01 03:00:00	3	0.002146	0.025745	0.705525	0.999998	0.999669	0.708685	0.962974	10.705371
2016-01-01 04:00:00	4	0.002861	0.034324	0.864532	0.999996	0.999411	0.502579	1.207768	10.496686