Related
I have a ecdf plot like this:
penguins = sns.load_dataset("penguins")
fig, ax = plt.subplots(figsize = (10,8))
sns.ecdfplot(data=penguins, x="bill_length_mm", hue="species")
ax.axhline(.25, linestyle = '--', color ='#cfcfcf', lw = 2, alpha = 0.75)
how to find the x values on this intersecting axhline?
You could loop through the generated curves (ax.get_lines()), extract their coordinates and search for the index of the first y-value larger than the desired y-value.
Here is some illustrating code (note that sns.ecdfplot() should get ax as parameter):
import matplotlib.pyplot as plt
import numpy as np
import seaborn as sns
penguins = sns.load_dataset("penguins")
fig, ax = plt.subplots(figsize=(10, 8))
sns.ecdfplot(data=penguins, x="bill_length_mm", hue="species", ax=ax)
y_special = 0.25
for line in ax.get_lines():
x, y = line.get_data()
ind = np.argwhere(y >= y_special)[0, 0] # first index where y is larger than y_special
# x[ind] is the desired x-value
ax.text(x[ind], y_special, f' {x[ind]:.1f}', ha='left', va='top') # maybe color=line.get_color()
ax.axhline(y_special, linestyle='--', color='#cfcfcf', lw=2, alpha=0.75)
plt.show()
PS: Optionally you could add these x-values to the legend:
for line, legend_text in zip(ax.get_lines(), ax.legend_.get_texts()):
x, y = line.get_data()
ind = np.argwhere(y >= y_special)[0, 0]
legend_text.set_text(f'{x[ind]:5.2f} {legend_text.get_text()}')
This is a case where it's better to use the computational tools that pandas provides instead of trying to back quantitative values out from a visual representation.
If you want the values corresponding to the .25 quantile for each species, you should do:
penguins.groupby("species")["bill_length_mm"].quantile(.25)
which returns
species
Adelie 36.75
Chinstrap 46.35
Gentoo 45.30
Name: bill_length_mm, dtype: float64
I am attempting to produce a plot like this which combines a cartesian scatter plot and a polar histogram. (Radial lines optional)
A similar solution (by Nicolas Legrand) exists for looking at differences in x and y (code here), but we need to look at ratios (i.e. x/y).
More specifically, this is useful when we want to look at the relative risk measure which is the ratio of two probabilities.
The scatter plot on it's own is obviously not a problem, but the polar histogram is more advanced.
The most promising lead I have found is this central example from the matplotlib gallery here
I have attempted to do this, but have run up against the limits of my matplotlib skills. Any efforts moving towards this goal would be great.
I'm sure that others will have better suggestions, but one method that gets something like you want (without the need for extra axes artists) is to use a polar projection with a scatter and bar chart together. Something like
import matplotlib.pyplot as plt
import numpy as np
x = np.random.uniform(size=100)
y = np.random.uniform(size=100)
r = np.sqrt(x**2 + y**2)
phi = np.arctan2(y, x)
h, b = np.histogram(phi, bins=np.linspace(0, np.pi/2, 21), density=True)
colors = plt.cm.Spectral(h / h.max())
ax = plt.subplot(111, projection='polar')
ax.scatter(phi, r, marker='.')
ax.bar(b[:-1], h, width=b[1:] - b[:-1],
align='edge', bottom=np.max(r) + 0.2, color=colors)
# Cut off at 90 degrees
ax.set_thetamax(90)
# Set the r grid to cover the scatter plot
ax.set_rgrids([0, 0.5, 1])
# Let's put a line at 1 assuming we want a ratio of some sort
ax.set_thetagrids([45], [1])
which will give
It is missing axes labels and some beautification, but it might be a place to start. I hope it is helpful.
You can use two axes on top of each other:
import matplotlib.pyplot as plt
fig = plt.figure(figsize=(6,6))
ax1 = fig.add_axes([0.1,0.1,.8,.8], label="cartesian")
ax2 = fig.add_axes([0.1,0.1,.8,.8], projection="polar", label="polar")
ax2.set_rorigin(-1)
ax2.set_thetamax(90)
plt.show()
Ok. Thanks to the answer from Nicolas, and the answer from tomjn I have a working solution :)
import numpy as np
import matplotlib.pyplot as plt
# Scatter data
n = 50
x = 0.3 + np.random.randn(n)*0.1
y = 0.4 + np.random.randn(n)*0.02
def radial_corner_plot(x, y, n_hist_bins=51):
"""Scatter plot with radial histogram of x/y ratios"""
# Axis setup
fig = plt.figure(figsize=(6,6))
ax1 = fig.add_axes([0.1,0.1,.6,.6], label="cartesian")
ax2 = fig.add_axes([0.1,0.1,.8,.8], projection="polar", label="polar")
ax2.set_rorigin(-20)
ax2.set_thetamax(90)
# define useful constant
offset_in_radians = np.pi/4
def rotate_hist_axis(ax):
"""rotate so that 0 degrees is pointing up and right"""
ax.set_theta_offset(offset_in_radians)
ax.set_thetamin(-45)
ax.set_thetamax(45)
return ax
# Convert scatter data to histogram data
r = np.sqrt(x**2 + y**2)
phi = np.arctan2(y, x)
h, b = np.histogram(phi,
bins=np.linspace(0, np.pi/2, n_hist_bins),
density=True)
# SCATTER PLOT -------------------------------------------------------
ax1.scatter(x,y)
ax1.set(xlim=[0, 1], ylim=[0, 1], xlabel="x", ylabel="y")
ax1.spines['right'].set_visible(False)
ax1.spines['top'].set_visible(False)
# HISTOGRAM ----------------------------------------------------------
ax2 = rotate_hist_axis(ax2)
# rotation of axis requires rotation in bin positions
b = b - offset_in_radians
# plot the histogram
bars = ax2.bar(b[:-1], h, width=b[1:] - b[:-1], align='edge')
def update_hist_ticks(ax, desired_ratios):
"""Update tick positions and corresponding tick labels"""
x = np.ones(len(desired_ratios))
y = 1/desired_ratios
phi = np.arctan2(y,x) - offset_in_radians
# define ticklabels
xticklabels = [str(round(float(label), 2)) for label in desired_ratios]
# apply updates
ax2.set(xticks=phi, xticklabels=xticklabels)
return ax
ax2 = update_hist_ticks(ax2, np.array([1/8, 1/4, 1/2, 1, 2, 4, 8]))
# just have radial grid lines
ax2.grid(which="major", axis="y")
# remove bin count labels
ax2.set_yticks([])
return (fig, [ax1, ax2])
fig, ax = radial_corner_plot(x, y)
Thanks for the pointers!
I'd like to draw a lognormal distribution of a given bar plot.
Here's the code
import matplotlib.pyplot as plt
from matplotlib.pyplot import figure
import numpy as np; np.random.seed(1)
import scipy.stats as stats
import math
inter = 33
x = np.logspace(-2, 1, num=3*inter+1)
yaxis = [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0.01,0.03,0.3,0.75,1.24,1.72,2.2,3.1,3.9,
4.3,4.9,5.3,5.6,5.87,5.96,6.01,5.83,5.42,4.97,4.60,4.15,3.66,3.07,2.58,2.19,1.90,1.54,1.24,1.08,0.85,0.73,
0.84,0.59,0.55,0.53,0.48,0.35,0.29,0.15,0.15,0.14,0.12,0.14,0.15,0.05,0.05,0.05,0.04,0.03,0.03,0.03, 0.02,
0.02,0.03,0.01,0.01,0.01,0.01,0.01,0.0,0.0,0.0,0.0,0.0,0.01,0,0]
fig, ax = plt.subplots()
ax.bar(x[:-1], yaxis, width=np.diff(x), align="center", ec='k', color='w')
ax.set_xscale('log')
plt.xlabel('Diameter (mm)', fontsize='12')
plt.ylabel('Percentage of Total Particles (%)', fontsize='12')
plt.ylim(0,8)
plt.xlim(0.01, 10)
fig.set_size_inches(12, 12)
plt.savefig("Test.png", dpi=300, bbox_inches='tight')
Resulting plot:
What I'm trying to do is to draw the Probability Density Function exactly like the one shown in red in the graph below:
An idea is to convert everything to logspace, with u = log10(x). Then draw the density histogram in there. And also calculate a kde in the same space. Everything gets drawn as y versus u. When we have u at a top twin axes, x can stay at the bottom. Both axes get aligned by setting the same xlims, but converted to logspace on the top axis. The top axis can be hidden to get the desired result.
import matplotlib.pyplot as plt
import numpy as np
from scipy import stats
inter = 33
u = np.linspace(-2, 1, num=3*inter+1)
x = 10**u
us = np.linspace(u[0], u[-1], 500)
yaxis = [0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0.01,0.03,0.3,0.75,1.24,1.72,2.2,3.1,3.9,
4.3,4.9,5.3,5.6,5.87,5.96,6.01,5.83,5.42,4.97,4.60,4.15,3.66,3.07,2.58,2.19,1.90,1.54,1.24,1.08,0.85,0.73,
0.84,0.59,0.55,0.53,0.48,0.35,0.29,0.15,0.15,0.14,0.12,0.14,0.15,0.05,0.05,0.05,0.04,0.03,0.03,0.03, 0.02,
0.02,0.03,0.01,0.01,0.01,0.01,0.01,0.0,0.0,0.0,0.0,0.0,0.01,0,0]
yaxis = np.array(yaxis)
# reconstruct data from the given frequencies
u_data = np.repeat((u[:-1] + u[1:]) / 2, (yaxis * 100).astype(np.int))
kde = stats.gaussian_kde((u[:-1]+u[1:])/2, weights=yaxis, bw_method=0.2)
total_area = (np.diff(u)*yaxis).sum() # total area of all bars; divide by this area to normalize
fig, ax = plt.subplots()
ax2 = ax.twiny()
ax2.bar(u[:-1], yaxis, width=np.diff(u), align="edge", ec='k', color='w', label='frequencies')
ax2.plot(us, total_area*kde(us), color='crimson', label='kde')
ax2.plot(us, total_area * stats.norm.pdf(us, u_data.mean(), u_data.std()), color='dodgerblue', label='lognormal')
ax2.legend()
ax.set_xscale('log')
ax.set_xlabel('Diameter (mm)', fontsize='12')
ax.set_ylabel('Percentage of Total Particles (%)', fontsize='12')
ax.set_ylim(0,8)
xlim = np.array([0.01,10])
ax.set_xlim(xlim)
ax2.set_xlim(np.log10(xlim))
ax2.set_xticks([]) # hide the ticks at the top
plt.tight_layout()
plt.show()
PS: Apparently this also can be achieved directly without explicitly using u (at the cost of being slightly more cryptic):
x = np.logspace(-2, 1, num=3*inter+1)
xs = np.logspace(-2, 1, 500)
total_area = (np.diff(np.log10(x))*yaxis).sum() # total area of all bars; divide by this area to normalize
kde = gaussian_kde((np.log10(x[:-1])+np.log10(x[1:]))/2, weights=yaxis, bw_method=0.2)
ax.bar(x[:-1], yaxis, width=np.diff(x), align="edge", ec='k', color='w')
ax.plot(xs, total_area*kde(np.log10(xs)), color='crimson')
ax.set_xscale('log')
Note that the bandwidth set for gaussian_kde is a somewhat arbitrarily value. Larger values give a more equalized curve, smaller values keep closer to the data. Some experimentation can help.
i have a little problem to create a subplot loop.
The following code show my result for one plot.... So it starts with a dayloop than with a hour loop (8 timesteps).
If i run the code i get a nice QUiver plot with the colorbar.
for dd in range(1,15):
day=str(dd)
readfile=fns[files_indizes[dd]]
if dd < 10:
nc_u_comp = NetCDFFile(ROOT+u_comp1+'0'+day+comp)
nc_v_comp = NetCDFFile(ROOT+v_comp1+'0'+day+comp)
else:
nc_u_comp = NetCDFFile(ROOT+u_comp1+day+comp)
nc_v_comp = NetCDFFile(ROOT+v_comp1+day+comp)
time = nc_u_comp.variables['time'][:]
index=readfile.find(comp)
index=index+len(comp)
date=readfile[index-14:index-6]
plt.clf()
for tt in range(0,len(time)):
if tt < 10:
h =str(0)+str(tt)
else:
h=str(tt)
varU=nc_u_comp.variables['u10'][tt,:,:]
varV=nc_v_comp.variables['v10'][tt,:,:]
lat = nc_u_comp.variables['latitude'][:]
lon = nc_u_comp.variables['longitude'][:]
plt.rcParams["figure.figsize"] = [10,10]
#plane projection of the world
#map with box size (defintion on the top)
box = sgeom.box(minx=llcrnrlon, maxx=urcrnrlon, miny=llcrnrlat, maxy=urcrnrlat)
x0, y0, x1, y1 = box.bounds
#Map plot. The middel of the map is central_longitude
#proj = ccrs.PlateCarree(central_longitude=0)
proj=ccrs.PlateCarree()
#Change middelpoint of the map
box_proj = ccrs.PlateCarree(central_longitude=0)
ax2 = plt.axes(projection=proj)
ax2.set_extent([x0, x1, y0, y1], box_proj)
ax2.add_feature(cartopy.feature.BORDERS, linestyle='-', alpha=.5)
ax2.coastlines(resolution='50m')
#Definition of the scale_bar
gl = ax2.gridlines(ccrs.PlateCarree(), \
linestyle='--', alpha=1, linewidth=0.5, draw_labels=True)
gl.xlabels_top = False
gl.ylabels_right = False
gl.xformatter = LONGITUDE_FORMATTER
gl.yformatter = LATITUDE_FORMATTER
magnitude = (varU ** 2 + varV ** 2) ** 0.5
strm =plt.streamplot(lon , lat , varU, varV, linewidth=2, density=2, color=magnitude)
cbar= plt.colorbar()
cbar.set_label('$m/s$')
name='Wind in 10 m '+ date + h+' UTC'
ax2.set_aspect('auto')
plt.title(name, y=1)
Now i want to create an 2x4 Subplot array with a colorbar allocate to the complete Subplot array.
I find some infromation in the internet, but it doesn't run with my code. Maybe someone can help me?
This shows how to plot an array of simple Cartopy maps in 4 rows 2 columns. Also shows how to plot a colorbar to accompany the maps array. Hope it helps.
import numpy as np
import cartopy.crs as ccrs
import matplotlib.pyplot as plt
import matplotlib as mpl
# create figure with figsize big enough to accomodate all maps, labels, etc.
fig = plt.figure(figsize=(8, 10), tight_layout=False)
# define plot array's arrangement
columns = 2
rows = 4
# set projection to use
projex = ccrs.PlateCarree()
# set the colormap and norm for
# the colorbar to use
cmap1 = mpl.cm.magma
norm1 = mpl.colors.Normalize(vmin=0, vmax=100)
def plotmymap(axs):
# your plot specs of each map should replace this
img = np.random.randint(100, size=(15, 30)) # 2d array of random values (1-100)
# render image on current axis
plims = plt.imshow(img, extent=[-180,180,-90,90], alpha=0.5, cmap=cmap1, norm=norm1)
axs.set_global()
axs.coastlines()
# add title to the map
axs.set_title("Map_"+str(i))
return plims # for use by colorbar
for i in range(1, columns*rows +1):
# add a subplot into the array of plots
ax = fig.add_subplot(rows, columns, i, projection=projex)
plims = plotmymap(ax) # a simple maps is created on subplot
# add a subplot for vertical colorbar
bottom, top = 0.1, 0.9
left, right = 0.1, 0.8
fig.subplots_adjust(top=top, bottom=bottom, left=left, right=right, hspace=0.15, wspace=0.25)
cbar_ax = fig.add_axes([0.85, bottom, 0.05, top-bottom])
fig.colorbar(plims, cax=cbar_ax) # plot colorbar
plt.show() # this plot all the maps
The resulting plots:
I am refering to the question already asked here.
In this example the users have solved the second axis problem by adding it to the upper part of the graph where it coincide with the title.
Question:
Is it possible to add the second x-axis at the bottom of the first one?
Code:
import numpy as np
import matplotlib.pyplot as plt
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax2 = ax1.twiny()
X = np.linspace(0,1,1000)
Y = np.cos(X*20)
ax1.plot(X,Y)
ax1.set_xlabel(r"Original x-axis: $X$")
new_tick_locations = np.array([.2, .5, .9])
def tick_function(X):
V = 1/(1+X)
return ["%.3f" % z for z in V]
ax2.set_xticks(new_tick_locations)
ax2.set_xticklabels(tick_function(new_tick_locations))
ax2.set_xlabel(r"Modified x-axis: $1/(1+X)$")
plt.show()
As an alternative to the answer from #DizietAsahi, you can use spines in a similar way to the matplotlib example posted here.
import numpy as np
import matplotlib.pyplot as plt
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax2 = ax1.twiny()
# Add some extra space for the second axis at the bottom
fig.subplots_adjust(bottom=0.2)
X = np.linspace(0,1,1000)
Y = np.cos(X*20)
ax1.plot(X,Y)
ax1.set_xlabel(r"Original x-axis: $X$")
new_tick_locations = np.array([.2, .5, .9])
def tick_function(X):
V = 1/(1+X)
return ["%.3f" % z for z in V]
# Move twinned axis ticks and label from top to bottom
ax2.xaxis.set_ticks_position("bottom")
ax2.xaxis.set_label_position("bottom")
# Offset the twin axis below the host
ax2.spines["bottom"].set_position(("axes", -0.15))
# Turn on the frame for the twin axis, but then hide all
# but the bottom spine
ax2.set_frame_on(True)
ax2.patch.set_visible(False)
# as #ali14 pointed out, for python3, use this
# for sp in ax2.spines.values():
# and for python2, use this
for sp in ax2.spines.itervalues():
sp.set_visible(False)
ax2.spines["bottom"].set_visible(True)
ax2.set_xticks(new_tick_locations)
ax2.set_xticklabels(tick_function(new_tick_locations))
ax2.set_xlabel(r"Modified x-axis: $1/(1+X)$")
plt.show()
I think you have to create a second Axes with 0 height (and hide the yaxis) to have a second xaxis that you can place wherever you like.
for example:
import numpy as np
import matplotlib.pyplot as plt
fig = plt.figure()
ax1 = fig.add_axes((0.1,0.3,0.8,0.6)) # create an Axes with some room below
X = np.linspace(0,1,1000)
Y = np.cos(X*20)
ax1.plot(X,Y)
ax1.set_xlabel(r"Original x-axis: $X$")
# create second Axes. Note the 0.0 height
ax2 = fig.add_axes((0.1,0.1,0.8,0.0))
ax2.yaxis.set_visible(False) # hide the yaxis
new_tick_locations = np.array([.2, .5, .9])
def tick_function(X):
V = 1/(1+X)
return ["%.3f" % z for z in V]
ax2.set_xticks(new_tick_locations)
ax2.set_xticklabels(tick_function(new_tick_locations))
ax2.set_xlabel(r"Modified x-axis: $1/(1+X)$")
plt.show()
Not really an answer to the question, but it took me quite long time until I figured out how to do the same with logscale. There are a bunch of strange behaviours in that case. Here's my code to apply some simple scaling to the original y axis:
def set_scaled_y_axis(ax, label1, label2, scale):
#define the minor and major ticks
#might give an error for too small or large exponents (e.g. 1e-20 or 1e+20)
log_ticks_major=[]
log_ticks_minor=[]
tick_labels=[]
for k in range(-15,16,1):
log_ticks_major.append(10**k)
tick_labels.append("10$^{"+f"{k}"+"}$")
for kk in range(2,10):
log_ticks_minor.append(kk*10**k)
log_ticks_major=np.array(log_ticks_major)
log_ticks_minor=np.array(log_ticks_minor)
#update the original label
ax.set_ylabel(label2)
# make a twin axis and set the position
# to make the same with x axis you need "ax.twiny()" instead
ax22 = ax.twinx()
ax22.yaxis.set_ticks_position("left")
ax22.yaxis.set_label_position("left")
ax22.spines["left"].set_position(("axes", -0.15))
# draw only the left y axis
ax22.xaxis.set_visible(False)
# set the log scale for the 2nd axis
ax22.set_yscale("log")
ax22.set_yticks(log_ticks_minor/scale, minor=True) # set minor ticks
ax22.set_yticks(log_ticks_major/scale) # set normal(/major?) ticks
ax22.set_yticklabels(tick_labels) #must be after "ax22.set_yticks(log_ticks_major/scale)"
ax22.tick_params('y', which="minor", labelleft=False) #some "random" minor tick labels would appear
# set the 2nd y axis label
ax22.set_ylabel(label1)
# set the limits of the 2nd y axis to be the same as the 1st one
ax22.set_ylim(ax.get_ylim())