Matplotlib Plot time series with colored points Pandas (Python)

Question

I'm trying to plot time series data where there are multiple points at each time. I want to have timeon the x axis, Amount on the y axis and then color each point by ID i.e. where ID = 344 they would all have the same color etc.

Below is example data i'm using. I'm sure this must exist but I can't find gallery examples on Matplotlib.org

                ID     Amount
Time                        
2015-12-09      344    0.333333
2015-12-10      345    0.333333
2015-12-09      345    0.333333
2015-12-09      344    0.750000
2015-12-09      342    0.583333

Things i've tried include trying to reshape the data as a pivot table (didn't work because there are two duplicate values for ID 344. Groupby, but I've struggled grouping by two columns, I think if I could groupby ID and retain the Time field this would go some way towards solving my problem.

Any help or advice would be massively appreciated.

score 5 · Accepted Answer · answered Jan 12 '16 at 20:53

5

Read the docs and look at examples for scatter within pylab or matplotlib

import pylab as pl

fig= pl.figure( figsize=(5,5) )
ax  = fig.add_subplot(111)

ax.scatter(df.index, df.Amount, s=20, c=df.ID)

This can be customized to meet your needs.

answered Jan 12 '16 at 20:53

tnknepp

5,888
6
43
57

Thanks for the suggestion, do you know if it's possible to use time stamp data for the x axis in a scatter plot? – hselbie Jan 12 '16 at 22:42
1

It is, but it must first be converted from string to datetime. – tnknepp Jan 14 '16 at 11:24

Matplotlib Plot time series with colored points Pandas (Python)

1 Answers1