Skip to content

Updated violin plot example as per suggestions in issue #7251 #7360

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
Oct 31, 2016
Merged

Updated violin plot example as per suggestions in issue #7251 #7360

Changes from 7 commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
08db92d
Updated example as per suggestions in issue #7251
DrNightmare Oct 29, 2016
becbfa4
Fixed ordering of parameters to np.clip()
DrNightmare Oct 29, 2016
14cd6cf
Made variable names more meaningful
DrNightmare Oct 30, 2016
03946d4
Replaced use of zip with call to transpose
DrNightmare Oct 30, 2016
8044e9d
Changed variable 'tmp' to 'quartiles' and slightly altered transpose …
DrNightmare Oct 30, 2016
88a8c2e
Fixed recalculation of interquartile ranges
DrNightmare Oct 30, 2016
62c1716
Replaced plot() calls with scatter() and vlines() calls
DrNightmare Oct 30, 2016
9f4bbdd
Removed unused variable and comments
DrNightmare Oct 31, 2016
b3ce73b
Removed unnecessary comments
DrNightmare Oct 31, 2016
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
109 changes: 43 additions & 66 deletions examples/statistics/customized_violin_demo.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,87 +18,64 @@
import numpy as np


# functions to calculate percentiles and adjacent values
def percentile(vals, p):
N = len(vals)
n = p*(N+1)
k = int(n)
d = n-k
if k <= 0:
return vals[0]
if k >= N:
return vals[N-1]
return vals[k-1] + d*(vals[k] - vals[k-1])


def adjacent_values(vals):
q1 = percentile(vals, 0.25)
q3 = percentile(vals, 0.75)
iqr = q3 - q1 # inter-quartile range

# upper adjacent values
uav = q3 + iqr * 1.5
if uav > vals[-1]:
uav = vals[-1]
if uav < q3:
uav = q3

# lower adjacent values
lav = q1 - iqr * 1.5
if lav < vals[0]:
lav = vals[0]
if lav > q1:
lav = q1
return [lav, uav]
def adjacent_values(vals, q1, q3):
upper_adjacent_value = q3 + (q3 - q1) * 1.5
upper_adjacent_value = np.clip(upper_adjacent_value, q3, vals[-1])

lower_adjacent_value = q1 - (q3 - q1) * 1.5
lower_adjacent_value = np.clip(lower_adjacent_value, vals[0], q1)
return [lower_adjacent_value, upper_adjacent_value]
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Probabky doesn't need to be bracketed [], wonder if it can be used to remove zip* stuff in the unpacking...

Copy link
Contributor Author

@DrNightmare DrNightmare Oct 31, 2016 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

To avoid the unpacking, I could also optionally make 'whiskers' an np.array and then do:
whiskersMin, whiskersMax = whiskers[:, 0], whiskers[:, 1]

Irrespective of whether I do this or not, the brackets can be removed

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Then I think it makes sense to do that.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, updated in latest commit



def set_axis_style(ax, labels):
ax.get_xaxis().set_tick_params(direction='out')
ax.xaxis.set_ticks_position('bottom')
ax.set_xticks(np.arange(1, len(labels) + 1))
ax.set_xticklabels(labels)
ax.set_xlim(0.25, len(labels) + 0.75)
ax.set_xlabel('Sample name')


# create test data
np.random.seed(123)
dat = [np.random.normal(0, std, 100) for std in range(1, 5)]
lab = ['A', 'B', 'C', 'D'] # labels
med = [] # medians
iqr = [] # inter-quantile ranges
avs = [] # upper and lower adjacent values
for arr in dat:
sarr = sorted(arr)
med.append(percentile(sarr, 0.5))
iqr.append([percentile(sarr, 0.25), percentile(sarr, 0.75)])
avs.append(adjacent_values(sarr))

# plot the violins
fig, (ax1, ax2) = plt.subplots(nrows=1, ncols=2, figsize=(9, 4),
sharey=True)
_ = ax1.violinplot(dat)
parts = ax2.violinplot(dat, showmeans=False, showmedians=False,
showextrema=False)
data = [sorted(np.random.normal(0, std, 100)) for std in range(1, 5)]

fig, (ax1, ax2) = plt.subplots(nrows=1, ncols=2, figsize=(9, 4), sharey=True)

# plot the default violin
ax1.set_title('Default violin plot')
ax2.set_title('Customized violin plot')
ax1.set_ylabel('Observed values')
ax1.violinplot(data)

# plot whiskers as thin lines, quartiles as fat lines,
# and medians as points
for i in range(len(med)):
# whiskers
ax2.plot([i + 1, i + 1], avs[i], '-', color='black', linewidth=1)
ax2.plot([i + 1, i + 1], iqr[i], '-', color='black', linewidth=5)
ax2.plot(i + 1, med[i], 'o', color='white',
markersize=6, markeredgecolor='none')
# customized violin
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

stray redundant comment

Copy link
Contributor Author

@DrNightmare DrNightmare Oct 31, 2016 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm, right. Guess the title we set for the axes are self explanatory.
Updated with comments removed @story645

ax2.set_title('Customized violin plot')
parts = ax2.violinplot(
data, showmeans=False, showmedians=False,
showextrema=False)

# customize colors
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

probably also unnecessary comment

for pc in parts['bodies']:
pc.set_facecolor('#D43F3A')
pc.set_edgecolor('black')
pc.set_alpha(1)

ax1.set_ylabel('Observed values')
quartile1, medians, quartile3 = np.percentile(data, [25, 50, 75], axis=1)
inter_quartile_ranges = np.vstack([quartile1, quartile3]).T
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pretty sure this variable is no longer used and so can be removed

whiskers = [
adjacent_values(sorted_array, q1, q3)
for sorted_array, q1, q3 in zip(data, quartile1, quartile3)]
whiskersMin, whiskersMax = list(zip(*whiskers))
# plot medians as points,
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can remove these comments, probably

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm right, will remove these comments and the unused variable

# whiskers as thin lines, quartiles as fat lines
inds = np.arange(1, len(medians) + 1)
ax2.scatter(inds, medians, marker='o', color='white', s=30, zorder=3)
ax2.vlines(inds, quartile1, quartile3, color='k', linestyle='-', lw=5)
ax2.vlines(inds, whiskersMin, whiskersMax, color='k', linestyle='-', lw=1)

# set style for the axes
labels = ['A', 'B', 'C', 'D'] # labels
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

variable name = comment

for ax in [ax1, ax2]:
ax.get_xaxis().set_tick_params(direction='out')
ax.xaxis.set_ticks_position('bottom')
ax.set_xticks(np.arange(1, len(lab) + 1))
ax.set_xticklabels(lab)
ax.set_xlim(0.25, len(lab) + 0.75)
ax.set_xlabel('Sample name')
set_axis_style(ax, labels)

plt.subplots_adjust(bottom=0.15, wspace=0.05)

plt.show()