current community

your communities

more stack exchange communities

Stack Exchange
tour help
careers 2.0
Take the 2-minute tour ×
Stack Overflow is a question and answer site for professional and enthusiast programmers. It's 100% free, no registration required.
up vote 0 down vote favorite

I am using pandas and matplotlib to plot data from an experiment involving 5 sessions. I would like the data for each session to be displayed in a separate panel; I am attempting to use subplotting to do this.

I nearly have what I want using the code below (see example figure here: http://imgur.com/99nK2TR). The problem is that different sessions have different numbers of trials and the scale is changed for different panels when I plot. I want the scale to be the same in all plots and the size of the subplot bounding box to adjust instead. Specifically, the trials should be equally spaced across all panels and consequently panels S1 and S2b (which contain 12 trials each) will be wider than panels S2a, S3a and S3b (which contain 3 trials each).

I think I might have to use some combination of aspect and/or adjustable and/or sharex but I can't fathom out how...

Apologies for the shoddy coding I'm new =)

fig,([ax1,ax2,ax3,ax4,ax5]) = plt.subplots(1, 5, sharey=True)

dfsubset1.plot(xticks=range(1,13,1), xlim=[0,13], ylim=[0,35], ax=ax1, title='S1', figsize=(12, 6), style='o-', legend=False)
dfsubset2.plot(xticks=range(1,4,1), xlim=[0,4], ylim=[0,35], ax=ax2, title='S2a', figsize=(12, 6), style='o-', legend=False)
dfsubset3.plot(xticks=range(1,13,1), xlim=[0,13], ylim=[0,35], ax=ax3, title='S2b', figsize=(12, 6), style='o-', legend=False)
dfsubset4.plot(xticks=range(1,4,1), xlim=[0,4], ylim=[0,35], ax=ax4, title='S3a', figsize=(12, 6), style='o-', legend=False)
dfsubset5.plot(xticks=range(1,4,1), xlim=[0,4], ylim=[0,35], ax=ax5, title='S3b', figsize=(12, 6), style='o-', legend=False)
share|improve this question

1 Answer 1

up vote 1 down vote

There are several ways to do this. Let's assume you have to read in your data into separate dataframes:

all_df = [df1, df2, ..., dfN]
fig, axes = plt.subplots(ncols=len(all_df), sharey=True)
for df, ax in zip(all_df, axes):
    df.plot(ax=ax)

You can put your x-limits and titles in lists (of lists) and zip them up in a loop too, though I highly recommend you compute them from the dataframe within the loop. No point in setting y-limits if sharey=True.

If you're actually able to have all of the data in individual columns of the same dataframe, it might be as simple as bigdf.plot(x='xcol', sharey=True, subplots=True)

share|improve this answer
    
Hi Paul thanks for your help. With hindsight I should have been clearer about the dataframe I am using - I tried to simplfy the example above but its a little more complicated so I'm struggling to implement your suggestion. –  user3408208 Mar 12 at 11:49
    
The df has a hierarchical index see here: imgur.com/ue2ZrbL When I try to follow your steps I'm just getting a series of vertical subplots all with the same length. The plots need to be horizontal - showing the timecourse of the experiment - and to the appropriate width (i.e. 12 trials in S1, 3 in S2a etc). Hope that's clearer... –  user3408208 Mar 12 at 11:56
    
here is something similar I made in ggplot2 a while ago: imgur.com/qMx1Bw9 NB in that case session 3a and 3b were collapsed. –  user3408208 Mar 12 at 12:23
    
@user3408208 you really need to post some of your data and a minimal working example. i still don't understand if it's all in one big data frame or not. here's an example of how to include your data in a minimal working examples in case you're not sure: stackoverflow.com/questions/22159805/… –  Paul H Mar 12 at 17:13

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.