Matplotlib df.plot() showing dots but not lines, even when specified

Question

I have a dataframe with a lot of missing values which looks like this:

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

date = pd.date_range(start='2003/01/01', end='2005/12/31')

df = pd.DataFrame({'date':date, })

Assign missing values to columns:

df = pd.DataFrame(np.nan, index=date, columns=['A', 'B'])

Add some actual values throughout to illustrate what my data actually looks like

df.loc['2003-01-10', 'B'] = 50
df.loc['2003-01-15', 'A'] = 70

df.loc['2003-06-10', 'B'] = 45
df.loc['2003-07-15', 'A'] = 55

df.loc['2004-01-01', 'B'] = 20
df.loc['2004-01-05', 'A'] = 30

df.loc['2004-05-01', 'B'] = 25
df.loc['2004-06-05', 'A'] = 35

df.loc['2005-01-01', 'B'] = 40
df.loc['2005-01-05', 'A'] = 35

Plot the data

df.plot(style = '-o')

This plot looks like this:

So you can see that I have specified that it be a line plot using the style = '-o' command, and it shows up correctly in the legend, but the dots are not joined by lines on the graph. When I plot it with no style specification I get a blank graph.

Any help would be greatly appreciated. Thank you.

Quickbeam2k1 · Accepted Answer · 2018-08-10 11:57:31Z

3

I assume this is due to the NaNs in your data set. Your data is simply not tidy. I assumed pandas could figure this out just using stack but it doesn't work either. Also, a bit inconvenient is that for a specific date not both values are defined( maybe one could use interpolate here. However, what works is simply:

df['A'].dropna().plot()
df['B'].dropna().plot()

in a single Jupiter notebook cell. Both plots will be drawn to the same axis there.

Interpolate works, but looks a bit different due to the scaling:

pd.concat([df['A'].interpolate(),
           df['B'].interpolate()], axis=1).plot()

note that here the legend is created directly. I was too lazy to overwrite the old df.

Tweaking interpolate a bit and realizing that it's already a DataFrame method one could also do:

df.interpolate(limit_area='inside').plot()

for qualitatively the drop_na result or

df.interpolate(limit_area='inside').plot()

for the concat result.

edited Aug 10, 2018 at 11:57

answered Aug 10, 2018 at 11:04

Quickbeam2k1

5,4573 gold badges31 silver badges43 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Pad Over a year ago

This works and your interpolate tip is great as it keeps the legend. Thank you :)

Quickbeam2k1 Over a year ago

just added some more content for interpolate.

ysearka · Accepted Answer · 2018-08-10 11:05:05Z

2

You have a lot of NaN values in your dataframe, so that it can't draw a line (the actual points aren't following each other).

What you can do is drop the nan values like this:

df.B.dropna().plot()
df.A.dropna().plot()

answered Aug 10, 2018 at 11:05

ysearka

3,8655 gold badges24 silver badges42 bronze badges

Collectives™ on Stack Overflow

Matplotlib df.plot() showing dots but not lines, even when specified

2 Answers 2

2 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related