how to work with adjustText with multiple subplots?

[zyxue]

I cannot make it work well with the following code patterns

1. adjust text for each subplot

fig, axes = plt.subplots(2, 2, figsize=(14, 12))
axes = axes.ravel()

for k, ax in enumerate(axes):
    # do some plotting
    texts = []
    # collect texts in a for loop
    # call adjust_text

2. adjust text for texts in all subplots together

fig, axes = plt.subplots(2, 2, figsize=(14, 12))
axes = axes.ravel()

texts = []
for k, ax in enumerate(axes):
    # do some plotting
    # collect texts in a for loop

# call adjust_text

What's the right way to do it? I haven't made it work in either case yet.

[Phlya]

Hi, The first option is correct. However make sure you also use the ax=ax argument in adjust_text. Let me know if this helps. Best, Ilya

…

On Thu, 19 Jul 2018, 19:42 Zhuyi Xue, ***@***.***> wrote: I cannot make it work well with the following code patterns 1. adjust text for each subplot fig, axes = plt.subplots(2, 2, figsize=(14, 12)) axes = axes.ravel() for k, ax in enumerate(axes): # do some plotting texts = [] # collect texts in a for loop # call adjust_text 1. adjust text for texts in all subplots together fig, axes = plt.subplots(2, 2, figsize=(14, 12)) axes = axes.ravel() texts = [] for k, ax in enumerate(axes): # do some plotting # collect texts in a for loop # call adjust_text What's the right way to do it? I haven't made it work in either case yet. — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#58>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACwsuq1PVeq9E8ktLsYe8gkbqozxdIcJks5uILbkgaJpZM4VWvAE> .

[zyxue]

Thanks, I will try it out. BTW, will ax.set_xlim or ax.set_ylim have an effect on adjustText? I saw some strange behavior before.

[Phlya]

Yes, always call adjust_text after changing axes limits.

…

On Thu, 19 Jul 2018, 22:35 Zhuyi Xue, ***@***.***> wrote: Thanks, I will try it out. BTW, will ax.set_xlim or ax.set_ylim have an effect on adjustText? I saw some strange behavior before. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#58 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ACwsusXfbHWpKo-7Xjdd60q_aykX_L15ks5uIN9ggaJpZM4VWvAE> .

[Phlya]

@zyxue did this solve your problems?

[zyxue]

@Phlya, thank you for asking, but no, it's not working yet. I am trying to get a minimal reproducible example

[zyxue]

I've made an example at https://github.com/zyxue/adjustText/blob/master/figures/Examples-for-multiple-subplots.ipynb.

Also, it seems to be much slower than in a single-subplot case (17.9 ms vs 13.4 s).

[Phlya]

As I mentioned above, you have to add ax=ax to adjust_text call:

fig, axes = plt.subplots(1, 2, figsize=(8, 3), sharex=True, sharey=True)
axes = axes.ravel()

for k, ax in enumerate(axes):
    np.random.seed(0)
    x, y = np.random.random((2,30))
    ax.plot(x, y, 'bo')

    texts = []
    for i in range(len(x)):
        t = ax.text(x[i], y[i], 'Text%s' %i, ha='center', va='center')
        texts.append(t)
    adjust_text(texts, ax=ax)

[zyxue]

Ah I see. I didn't fully get that. Ok, I will try it out later and let you know. Thank you!

[Phlya]

Concerning timing, make sure you are using axes of the same size/shape, when comparing single vs multiple subplots. If they are slightly smaller in one case, it can slow down the execution considerably.

[zyxue]

I don't fully understand what you mean by "If they are slightly smaller in one case, it can slow down the execution considerably".

I have updated my notebooks, why the time spent is so different for adjusting the first and second subplots, please?

CPU times: user 11.7 s, sys: 236 ms, total: 12 s
Wall time: 12 s
CPU times: user 2.52 s, sys: 57.8 ms, total: 2.57 s
Wall time: 2.58 s

BTW, are you interested in a pull request, I thought having a such an example for multiple subplots would be helpful to other users. It's a very useful package, thank you for sharing it.

[Phlya]

Oh, I see. Interesting, I don't know why that is so different, for some reason it does a very different number of iterations each time! I'll try to investigate when I have time...

[Phlya]

Yeah, would be great to add an example about this, you are not the first person to ask this question!

[zyxue]

ok, I've sent a pull request. Not I also removed figures/.ipynb_checkpoints from the commit as it's in your .gitignore already.

[zyxue]

Reopen because the time discrepancy for different subplots hasn't been resolsoved

Quote a previous comment:

CPU times: user 11.7 s, sys: 236 ms, total: 12 s
Wall time: 12 s
CPU times: user 2.52 s, sys: 57.8 ms, total: 2.57 s
Wall time: 2.58 s