Fix range of matplotlib histogram with multiple datasets?

byMR

August 4, 2023

I have multiple datasets of that I’d like to combine into a histogram. Unfortunately, my data shows different ranges.

Using the simple range=(0,1) argument triggers weird behaviour.

This is what it should look like:

and it currently does when the data has the range 0-1.

But sometimes it has not and I get:

Here I would like to force the 0-1 range.

But when I do with

axs.hist(plot_data, range=(0,1),color=colors,label=plot_label, histtype='bar')

I get

I cannot explain why that happens.

This is the code:

fig, axs = plt.subplots(1,1,tight_layout=True)
fig.set_size_inches(7, 4)
# Define colors for each histogram
colors = ['g', 'b', 'r', 'purple']
# make combined histogram for all error conditions
# plot the histogram
axs.hist(plot_data,color=colors,label=plot_label, histtype='bar')
# Put a legend to the right of the current axis
axs.legend(loc='center left',bbox_to_anchor=(1, 0.5))
# set the axis labels
axs.set_xlabel(f"Similarity: {wildcards.metric}")
axs.set_ylabel("Frequency")
# have the x-axis go increasing from left to right
axs.invert_xaxis()
# ensure proper layout
fig.tight_layout()
# save the histogram
fig.savefig(Path(output.hist))

and some example data is:

[array([
       '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0','1.0',
       '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0',
       '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0',
       '1.0', '1.0', '1.0'], dtype='<U46'),
 array(['
       1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0',
       '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0',
       '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0', '1.0',
       '1.0', '1.0', '1.0'], dtype='<U46'),
     array([
      '0.25', '0.3', '0.3', '0.3', '0.3', '0.3', '0.0', '0.0', '0.3',
       '0.0', '0.3', '0.3', '0.0', '0.0', '0.3', '0.5', '0.4', '0.0',
       '0.5', '0.5', '0.5', '0.0', '0.4', '0.0', '0.0', '0.6', '0.3',
       '0.1', '0.0', '0.0'], dtype='<U46'),
     array([
       '0.0', '0.0', '0.0', '0.0', '0.0', '0.0', '0.0','0.0','0.0',
       '0.0', '0.0', '0.0', '0.0', '0.0', '0.0', '0.0', '0.0', '0.0',
       '0.0', '0.0', '0.0', '0.0', '0.0', '0.0', '0.0', '0.0', '0.0',
       '0.0', '0.0', '0.0'], dtype='<U46')]

I have tried every possible range and bin combination.

>Solution :

Your arrays are full of string values but you’re passing a range parameter that is expecting float/int values. I’m not sure why your values are strings, but using float values (and not using dtype = '<U46') leads to the graph behaving as expected with the range parameter:

plot_data = [np.array([
       1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0,1.0,
       1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0,
       1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0,
       1.0, 1.0, 1.0]),
 np.array([
     1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0,
     1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0,
     1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0, 1.0,
     1.0, 1.0, 1.0]),
 np.array([
     0.25, 0.3, 0.3, 0.3, 0.3, 0.3, 0.0, 0.0, 0.3,
     0.0, 0.3, 0.3, 0.0, 0.0, 0.3, 0.5, 0.4, 0.0,
     0.5, 0.5, 0.5, 0.0, 0.4, 0.0, 0.0, 0.6, 0.3,
     0.1, 0.0, 0.0]),
 np.array([
     0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,0.0,0.0,
     0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
     0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0,
     0.0, 0.0, 0.0])]

fig, axs = plt.subplots(1,1,tight_layout=True)
fig.set_size_inches(7, 4)
# Define colors for each histogram
colors = ['g', 'b', 'r', 'purple']
plot_label = ['IDEAL', 'TYPICAL', 'LARGE', 'EXTREME']
# make combined histogram for all error conditions
# plot the histogram
axs.hist(plot_data, range=(0,1),color=colors,label=plot_label, histtype='bar')
#axs.hist(plot_data,histtype='bar', color=colors, label=plot_label)
# Put a legend to the right of the current axis
axs.legend(loc='center left',bbox_to_anchor=(1, 0.5))
# set the axis labels
axs.set_xlabel(f"Similarity:")
axs.set_ylabel("Frequency")
# have the x-axis go increasing from left to right
axs.invert_xaxis()
# ensure proper layout
fig.tight_layout()
plt.xlim(0, 1)
plt.show()