subplots in matplotlib using histogram

Question

diff  = [[10,20,30],[40,50,60],[70,80,90]]
comp = ["foo","bar","baz"]
fig,ax = plt.subplots()
for foo in range(0, len(diff)):
        x = [diff[foo]]
        name = comp
        color = ['0.1', '0.2', '0.3']
        label = ['1000000','1200000', '1400000']
        y = zip(*x)
        pos = np.arange(len(x))
        width = 1. / (1 + len(x))

        fig = plt.subplot(3,1,foo)
        for idx, (serie, color,label) in enumerate(zip(y, color,label)):
                ax.bar(pos + idx * width, serie, width, color=color,label=label)
        fig = plt.gcf()
        fig.set_size_inches(28.5,10.5)

        ax.set_xticks(pos + 1.5*width)
        plt.ylabel(name[foo])
        ax.set_xticklabels(comp)
        ax.legend()
        plt.gray()
plt.savefig("file" + '.jpg', bbox_inches='tight', pad_inches=0.5,dpi=100)
plt.clf()

I want to subplot foo bar and baz. But when I try to do that using the above code. The data is not being displayed on the graph. Any idea why?

I found a very nasty shortcut, which I don't like. Using Montage to join the images. :) Something like this montage -geometry 500 image1 image2 [...] output — pistal
– pistal, Commented Nov 20, 2013 at 14:52
You've called subplots twice: once before the loop with no arguments, which creates a figure with just one axis, but then in the loop you call it again with (3,1, ...) You want to move that second call to the top and then plot once in each of the three axes. — askewchan
– askewchan, Commented Nov 20, 2013 at 15:00

Saullo G. P. Castro · Accepted Answer · 2013-11-20 18:42:25Z

2

You are replacing the first fig when you call subplot inside the loop, here is a fixed version. See that ax returned by subplots is a np.ndarray, so you have to give an index ax[foo] to obtain the AxesSubplot object.

diff  = [[10, 20, 30], [40, 50, 60], [70, 80, 90]]
comp = ["foo", "bar", "baz"]
fig, ax = plt.subplots(3, 1)
for foo in range(0, len(diff)):
        x = [diff[foo]]
        name = comp
        color = ['0.1', '0.2', '0.3']
        label = ['1000000', '1200000', '1400000']
        y = zip(*x)
        pos = np.arange(len(x))
        width = 1. / (1 + len(x))
        for idx, (serie, color,label) in enumerate(zip(y, color,label)):
                ax[foo].bar(pos + idx * width, serie, width, color=color,label=label)
        fig.set_size_inches(28.5, 10.5)
        ax[foo].set_xticks(pos + 1.5*width)
        plt.ylabel(name[foo])
        ax[foo].set_xticklabels(comp)
        ax[foo].legend()
        plt.gray()
fig.savefig("file" + '.jpg', bbox_inches='tight', pad_inches=0.5, dpi=100)
plt.clf()

edited Nov 20, 2013 at 18:42

answered Nov 20, 2013 at 15:02

Saullo G. P. Castro

59.4k28 gold badges191 silver badges244 bronze badges

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

2 Comments

tacaswell Over a year ago

Why do you call gcf() and resize the figure every time through the loop? You should use fig.savefig(...) instead of plt.savefig

Saullo G. P. Castro Over a year ago

@tcaswell thank you! I've just copied the code from the OP and fixed the problem but did not pay enough attention...

Collectives™ on Stack Overflow

subplots in matplotlib using histogram

1 Answer 1

2 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related