Vary xytext to prevent overlapping annotations - python

So I've got some code that generates a donut chart but the problem is there are cases where the annotations overlap due to the values. Code and problem below.
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
def donut_chart(val):
df_vals = pd.DataFrame.from_dict(val, orient='index')
labels = df_vals.index.tolist()
fig, ax = plt.subplots(figsize=(6, 6), subplot_kw=dict(aspect="equal"))
color = ['grey']*20
color[0] = 'red'
wedges, texts, junk = ax.pie(df_vals[0:4], counterclock = True,
wedgeprops=dict(width=0.6, linewidth = 2, edgecolor = 'w'),
startangle=90, colors=color,
autopct='%1.0f%%',
pctdistance=0.75,
textprops={'fontsize': 14})
bbox_props = dict(boxstyle="square,pad=0.3", fc="w", ec="w", lw=0.72)
kw = dict(xycoords='data', textcoords='data', arrowprops=dict(arrowstyle="-"),
bbox=bbox_props, zorder=0, va="center")
for i, p in enumerate(wedges):
ang = (p.theta2 - p.theta1)/2. + p.theta1
y = np.sin(np.deg2rad(ang))
x = np.cos(np.deg2rad(ang))
horizontalalignment = {-1: "right", 1: "left"}[int(np.sign(x))]
connectionstyle = "angle,angleA=0,angleB={}".format(int(ang))
kw["arrowprops"].update({"connectionstyle": connectionstyle})
ax.annotate(labels[i], xy=(x, y), xytext=(1.2*np.sign(x), 1.2*y),
horizontalalignment=horizontalalignment, **kw, size=14)
#centre_circle = plt.Circle((0,0),0.5, fc='white',linewidth=1.25)
#fig.gca().add_artist(centre_circle)
plt.axis('equal')
plt.show()
plt.close()
val = {'Label A':50, 'Label B':2, 'Label C':1, 'Label D':0.5}
donut_chart(val)
Problem:
What I'd like to do is create something like this:
The key appears to be varying the y value in the xytext so the labels don't overlap but I'm stuck on how this might be implemented or even whether it is possible.
Any ideas?

updated code
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
def donut_chart(val):
df_vals = pd.DataFrame.from_dict(val, orient='index')
labels = df_vals.index.tolist()
fig, ax = plt.subplots(figsize=(6, 6), subplot_kw=dict(aspect="equal"))
color = ['grey']*20
color[0] = 'red'
wedges, texts = ax.pie(df_vals[0:5], counterclock = True,
wedgeprops=dict(width=0.6, linewidth = 1, edgecolor = 'w'),
startangle=90, colors=color,
textprops={'fontsize': 14})
bbox_props = dict(boxstyle="square,pad=0", fc="w", ec="w", lw=0.72)
kw = dict(xycoords='data', textcoords='data', arrowprops=dict(arrowstyle="-"),
bbox=bbox_props, zorder=0, va="center")
for i, p in enumerate(wedges):
ang = (p.theta2 - p.theta1)/2. + p.theta1
y = np.sin(np.deg2rad(ang))
x = np.cos(np.deg2rad(ang))
horizontalalignment = {-1: "right", 1: "left"}[int(np.sign(x))]
connectionstyle = "angle,angleA=0,angleB={}".format(int(ang))
kw["arrowprops"].update({"connectionstyle": connectionstyle})
ax.annotate(labels[i], xy=(x, y), xytext=((i/10) + 1.1*np.sign(x), (i/10) + y),
horizontalalignment=horizontalalignment, **kw, size=14)
#centre_circle = plt.Circle((0,0),0.5, fc='white',linewidth=1.25)
#fig.gca().add_artist(centre_circle)
plt.axis('equal')
plt.show()
plt.close()
val = {'Label A':50, 'Label B':2, 'Label C':0.2, 'Label D':0.2,'Label E':0.2}
donut_chart(val)
difference
ax.annotate(labels[i], xy=(x, y), xytext=((i/10) + 1.1*np.sign(x), (i/10) + y),
horizontalalignment=horizontalalignment, **kw, size=14)
output

Related

How to insert a small square mark somewhere on a generated heatmap plot

I am creating a 2D matplotlib plot (i and j coordinates) which contains 10 subplots. Each subplot contains 150 by 150 grid cell data. How can I insert a small black-colored square mark (3 by 3 ) somewhere fixed (center at coordinates 62 and 62 ) on each generated heatmap sub-plot across those 10 sub-plots? The square mark would therefore contain 10 blocks from 60 to 64 in both x and y direction and contains a written text "Sale 1" centered at x 62 and y 62. My code below does not generate any patches. Any feedback is greatly appreciated.
from matplotlib.patches import Rectangle
import numpy as np
import matplotlib.pyplot as plt
from sklearn.metrics import r2_score, median_absolute_error
import os
import matplotlib.cm as cm
from mpl_toolkits import axes_grid1
import matplotlib.pyplot as plt
#import seaborn as sns
import matplotlib.pyplot as plt
import matplotlib.colors as mcolors
import matplotlib.colors
import matplotlib.colors as colors
data = np.random.rand(10, 150, 150)
data = data.reshape(-1, 1)
property = "Sale"
pmin = data.min()
pmax = data.max()
v = np.linspace(round(pmin,3), round(pmax,3),15, endpoint=True)
v = [round(x,3) for x in v]
fig, ax = plt.subplots(2, 5, figsize=(160, 80))
row_count = 0
col_count = 0
for i in range(10):
sub_plot_data = data[(i)*(150*150):(i+1)*150*150]
x = 150
y = 150
#--------------------------- Define the map boundary ----------------------
xmin = 1258096.6
xmax = 1291155.0
ymin = 11251941.6
ymax = 11285000.0
pmin = min(sub_plot_data)
pmax = max(sub_plot_data)
# --------------------------- define color bar for Discrete color
bounds = np.linspace(-1, 1, 10)
Discrete_colors = plt.get_cmap('jet')(np.linspace(0,1,len(bounds)+1))
# create colormap without the outmost colors
cmap = mcolors.ListedColormap(Discrete_colors[1:-1]) #
actual_2d = np.reshape(sub_plot_data,(y,x))
im1 = ax[row_count, col_count].imshow(actual_2d, interpolation=None, cmap=cmap,
extent=(xmin, xmax, ymin, ymax), vmin=pmin, vmax=pmax)
plt.text(actual_2d[62, 62], actual_2d[62, 62], '%s' % 'Sale_1',
horizontalalignment='center', verticalalignment='center', color= 'black', fontsize= 90)
ax[row_count, col_count].set_title("Sale_Stores-%s - L: %s"%(i+1, layer),
fontsize=130, pad=44, x=0.5, y=0.999) # new
ax[row_count, col_count].set_aspect('auto')
ax[row_count, col_count].tick_params(left=False, labelleft=False, top=False,
labeltop=False, right=False, labelright=False, bottom=False, labelbottom=False) # new
#ax[row_count, col_count] = plt.gca()
plt.gca().add_patch(Rectangle((60, 60), 3, 3, edgecolor='black',
facecolor='black',fill=True,lw=2))
ax[row_count, col_count].add_patch(plt.text(62, 62, '%s' % 'Sale_1',
horizontalalignment='center', verticalalignment='center', color= 'black', fontsize= 90))
col_count +=1
if col_count == 5:
row_count +=1
col_count =0
fig.tight_layout(h_pad=10)
plt.subplots_adjust(left=0.02,
bottom=0.1,
right=0.91,
top=0.8,
wspace=0.1,
hspace=0.2)
cbaxes = fig.add_axes([0.94, 0.05, 0.02, 0.8])
cbar = fig.colorbar(im1, ax=ax.ravel().tolist(), ticks=v, extend='both', cax =cbaxes)
cbar.ax.tick_params(labelsize=70)
#cbar.set_ticks(v)
cbar.ax.set_yticklabels([i for i in v], fontsize=120)
output_dir = r"D/test"
plot_dir = os.path.join(output_dir, reservoir_property)
if not os.path.exists(plot_dir):
os.makedirs(plot_dir)
fig.savefig(r"%s/per_allmodel.png"%(plot_dir))
I tried your code and made a couple of modifications: first, the graph size was too huge and caused errors, so I made it smaller; second, I simplified the subplots: axes has a list of subplot objects, so I took them out with axes.flat; third The second is modifying the text as annotations. The graph size has been reduced and the font size and spacing have been adjusted, so please modify it yourself. Finally, tick_params is not set since the color bar ticks are disabled.
fig, axes = plt.subplots(2, 5, figsize=(16, 8))
row_count = 0
col_count = 0
for i,ax in enumerate(axes.flat):
sub_plot_data = data[(i)*(150*150):(i+1)*150*150]
x = 150
y = 150
#--------------------------- Define the map boundary ----------------------
xmin = 1258096.6
xmax = 1291155.0
ymin = 11251941.6
ymax = 11285000.0
pmin = min(sub_plot_data)
pmax = max(sub_plot_data)
# --------------------------- define color bar for Discrete color
bounds = np.linspace(-1, 1, 10)
Discrete_colors = plt.get_cmap('jet')(np.linspace(0,1,len(bounds)+1))
# create colormap without the outmost colors
cmap = mcolors.ListedColormap(Discrete_colors[1:-1]) #
actual_2d = np.reshape(sub_plot_data,(y,x))
#im = ax.imshow(actual_2d, interpolation=None, cmap=cmap, extent=(xmin, xmax, ymin, ymax), vmin=pmin, vmax=pmax)
im = ax.imshow(actual_2d, interpolation=None, cmap=cmap)
ax.text(actual_2d[62, 62], actual_2d[62, 62]-10, '%s' % 'Sale_1',
horizontalalignment='center', verticalalignment='center', color= 'black', fontsize=18)
ax.set_title("Sale_Stores-%s - L: %s"%(i+1, 1), fontsize=14, pad=30, x=0.5, y=0.999)
ax.set_aspect('auto')
ax.add_patch(Rectangle((60, 60), 6, 6, edgecolor='red', facecolor='red', fill=True, lw=2))
ax.text(62, 62, '%s' % 'Sale_1', ha='center', va='center', color='black', fontsize=14)
fig.tight_layout(h_pad=10)
plt.subplots_adjust(left=0.02,
bottom=0.1,
right=0.91,
top=0.8,
wspace=0.1,
hspace=0.5)
cbaxes = fig.add_axes([0.94, 0.05, 0.02, 0.8])
cbar = fig.colorbar(im, ax=axes.flat, ticks=v, extend='both', cax=cbaxes)
cbar.ax.tick_params(labelsize=10)
#cbar.set_ticks(v)
cbar.ax.set_yticklabels([str(i) for i in v], fontsize=12)
#plt.tick_params(left=False, labelleft=False, top=False, labeltop=False, right=False, labelright=False, bottom=False, labelbottom=False)
plt.show()

Gradient fill from zero till a curve

I have been using Is it possible to get color gradients under curve in matplotlib? as a reference (you can see the similarities, however i cant for the life of me figure out how to push the shading all the way down to 0 on the Y AXIS, for some reason which i cant find out, it has an upward sloping straight line cutting off the shading, i cant find anything in my data to suggest why its doing this.
for context the y axis can show positive and negative and i want to fill the scale the whole way so using gradient colour to fill from 0 to the line (positive) then fill from 0 to the negative line (see my blue example from a previous chart -same data-)
Here is my code
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.colors as mcolors
from matplotlib.patches import Polygon
# Variables
AUM = df['#AHD_AUM'].head(104)
MM = df['#AHD_Managed_Money_Net'].head(104)
PRICE = df['#AHD_Price'].head(104)
DATES = df['DATES'].head(104)
# Date Friendly Variables for Plot
List_AUM = df['#AHD_AUM'].head(104).to_list()
List_MM = df['#AHD_Managed_Money_Net'].head(104).to_list()
List_DATES = df['DATES'].head(104).to_list()
X = 0 * df['#AHD_AUM'].head(104)
# Make a date list changing dates with numbers to avoid the issue with the plot
interpreting dates
for i in range(len(df['DATES'].head(104))):
count = i
df['count'][i] = 120 - i
# X and Y data variables changed to arrays as when i had these set as dates
matplotlib hates it
x = df['count'].head(104).to_numpy()
y = df['#AHD_Managed_Money_Net'].head(104).to_numpy()
#DD = AUM.to_numpy()
#MMM = MM.to_numpy()
def main():
for _ in range(len(DD)):
gradient_fill(x,y)
plt.show()
def gradient_fill(x,y, fill_color=None, ax=None, **kwargs):
"""
"""
if ax is None:
ax = plt.gca()
line, = ax.plot(x, y, **kwargs)
if fill_color is None:
fill_color = line.get_color()
zorder = line.get_zorder()
alpha = line.get_alpha()
alpha = 1.0 if alpha is None else alpha
z = np.empty((100, 1, 4), dtype=float)
rgb = mcolors.colorConverter.to_rgb(fill_color)
z[:,:,:3] = rgb
z[:,:,-1] = np.linspace(0, alpha, 100)[:,None]
xmin, xmax, ymin, ymax = x.min(), x.max(), y.min(), y.max()
im = ax.imshow(z, aspect='auto', extent=[xmin, xmax, ymin, ymax],
origin='lower', zorder=zorder)
xy = np.column_stack([x, y])
# xy = np.vstack([[xmin, ymin], xy, [xmax, ymin], [xmin, ymin]]) ### i dont
need this so i have just commented it out
clip_path = Polygon(xy, facecolor='none', edgecolor='none', closed=True)
ax.add_patch(clip_path)
im.set_clip_path(clip_path)
ax.autoscale(True)
return line, im
main()
this is my current output
An easier way to clip the gradient by the curve, is to use a polygon obtained from fill_between.
Here is some example code to get you started.
import numpy as np
import matplotlib.pyplot as plt
np.random.seed(123)
x = np.linspace(0, 10, 200)
y = np.random.normal(0.01, 1, 200).cumsum()
fig, ax = plt.subplots(figsize=(12, 5))
ax.plot(x, y)
ylim = ax.get_ylim()
grad1 = ax.imshow(np.linspace(0, 1, 256).reshape(-1, 1), cmap='Blues', vmin=-0.5, aspect='auto',
extent=[x.min(), x.max(), 0, y.max()], origin='lower')
poly_pos = ax.fill_between(x, y.min(), y, alpha=0.1)
grad1.set_clip_path(poly_pos.get_paths()[0], transform=ax.transData)
poly_pos.remove()
grad2 = ax.imshow(np.linspace(0, 1, 256).reshape(-1, 1), cmap='Reds', vmin=-0.5, aspect='auto',
extent=[x.min(), x.max(), y.min(), 0], origin='upper')
poly_neg = ax.fill_between(x, y, y.max(), alpha=0.1)
grad2.set_clip_path(poly_neg.get_paths()[0], transform=ax.transData)
poly_neg.remove()
ax.set_ylim(ylim)
ax.axhline(0, color='black') # show a line at x=0
plt.show()
PS: vmin in imshow can be used to remove the color range where it's very light:
grad1 = ax.imshow(np.linspace(0, 1, 256).reshape(-1, 1), cmap='Blues', vmin=-0.5, aspect='auto',
extent=[x.min(), x.max(), 0, y.max()], origin='lower')
grad2 = ax.imshow(np.linspace(0, 1, 256).reshape(-1, 1), cmap='Reds', vmin=-0.5, aspect='auto',
extent=[x.min(), x.max(), y.min(), 0], origin='upper')
import pandas as pd # For data handling
import seaborn as sns # For plotting
import numpy as np
import matplotlib.pyplot as plt # For plotting
import matplotlib
#some preferred user settings
plt.rcParams['figure.figsize'] = (18.0, 12.0)
pd.set_option('display.max_columns', None)
%matplotlib inline
import warnings
warnings.filterwarnings(action='ignore')
from mpl_toolkits.axisartist.parasite_axes import HostAxes, ParasiteAxes
import matplotlib.pyplot as plt
from matplotlib.ticker import MultipleLocator
import datetime as dt
import matplotlib.dates as mdates
import pandas
Metal = CAD
# Variables
AUM = Metal.iloc[:,[7]].head(104)
MM = Metal.iloc[:,[0]].head(104)
PRICE = Metal.iloc[:,[8]].head(104)
#Last_Report = Metal.iloc[:,[9]].head(1).dt.strftime('%d %b %Y').to_list()
DATES = Metal.iloc[:,[10]].head(104)
# Dataframe for Net Position High
Net_High = Metal[Metal.iloc[:,[0]] == Metal.iloc[:,[0]].max()]
# Variables for Chart Annotation for Net Position High
Pos_High_Date = Net_High.iloc[:, [0]]
Pos_High_AUM = Net_High.iloc[:, [7]][0]/[1000000000]
Pos_High_Price = Net_High.iloc[:, [8]].to_numpy()[0].round().astype('int')
Pos_High = Net_High.iloc[:, [0]][0].astype('int')
Str_Date = mdates.num2date(Pos_High_Date)
Str_Date = pd.to_datetime(Str_Date[0]).strftime("%d %b %y")[0]
# Dataframe for Net Position Low
Net_Low = df[df['#CAD_Managed_Money_Net'] == df['#CAD_Managed_Money_Net'].head(104).min()]
# Variables for Chart Annotation for Net Position High
Pos_Low_Date = Net_Low.iloc[:, [55]].to_numpy()
Pos_Low_AUM = Net_Low.iloc[:, [26]].to_numpy()[0].round()/[1000000000]
Pos_Low_Price = Net_Low.iloc[:, [27]].to_numpy()[0].round().astype('int')
Pos_Low = Net_Low['#CAD_Managed_Money_Net'][0].astype('int')
Str_Date_Low = mdates.num2date(Pos_Low_Date)
Str_Date_Low = pd.to_datetime(Str_Date_Low[0]).strftime("%d %b %y")[0]
# C Brand Colour Scheme
C = ['deepskyblue', '#003399', 'slategray', '#027608','#cc0000']
def make_patch_spines_invisible(ax):
ax.set_frame_on(True)
ax.patch.set_visible(False)
for sp in ax.spines.values():
sp.set_visible(False)
fig, host = plt.subplots(figsize=(25,15))
fig.subplots_adjust(right=0.8)
#twinx() creates another axes sharing the x axis we do this twice
par1 = host.twinx()
par2 = host.twinx()
# Offset the right spine of par2 the ticks
par2.spines["right"].set_position(("axes",1.08))
#because par2 was created by twinx the frame is off so we need to use the method created above
make_patch_spines_invisible(par2)
# second, show the right spine
par2.spines["right"].set_visible(True)
######### Colouring in Plots
x = DATES
y = MM
ylim = host.get_ylim()
Long = host.imshow(np.linspace(0, 1, 256).reshape(-1, 1), cmap= 'Blues', vmin=-0.5, aspect='auto',
extent=[x.min(), x.max(), 0, y.max()], origin='lower')
poly_pos = host.fill_between(x, y.min(), y, alpha=0.1)
Long.set_clip_path(poly_pos.get_paths()[0], transform=host.transData)
poly_pos.remove()
Short = host.imshow(np.linspace(0, 1, 256).reshape(-1, 1), cmap='OrRd', vmin=-0.5, aspect='auto',
extent=[x.min(), x.max(), y.min(), 0], origin='upper')
poly_neg = host.fill_between(x, y, y.max(), alpha=0.1)
Short.set_clip_path(poly_neg.get_paths()[0], transform=host.transData)
poly_neg.remove()
##########
#plot data
p1, = host.plot(DATES, MM, label="Managed Money Net Position", linewidth=0.0,color = Citi[1], alpha = 0.8)
p2, = par1.plot(DATES, AUM, label="AUM",linewidth=1, marker = '$A$',mew = 1,mfc = 'w', color = Citi[0], alpha = 0.8)
p3, = par2.plot(DATES, PRICE, label="3M Price",linewidth=1, marker = '$p$', color = Citi[2], alpha = 0.8)
#Automatically scale and format
host_labels = ['{:,.0f}'.format(x) + 'K Lots' for x in host.get_yticks()/1000]
host.set_yticklabels(host_labels)
par1_labels = ['{:,.1f}'.format(x) + ' $Billion' for x in par1.get_yticks()/1000000000]
par1.set_yticklabels(par1_labels)
par2_labels = ['{:,.0f}'.format(x) + ' $' for x in par2.get_yticks()]
par2.set_yticklabels(par2_labels)
# x Axis formatting (date)
formatter = matplotlib.dates.DateFormatter('%b- %Y')
host.xaxis.set_major_formatter(formatter)
# Rotates and right-aligns the x labels so they don't crowd each other.
for label in host.get_xticklabels(which='major'):
label.set(rotation=30, horizontalalignment='right')
# Axis Labels
host.set_xlabel("Date")
host.set_ylabel("Managed Money Net Position")
par1.set_ylabel("AUM")
par2.set_ylabel("3M Price")
# Tick Parameters
tkw = dict(size=10, width=2.5)
# Set tick colours
host.tick_params(axis = 'y', colors = Citi[1], **tkw)
par1.tick_params(axis = 'y', colors = Citi[0], **tkw)
par2.tick_params(axis = 'y', colors = Citi[2], **tkw)
#host.tick_params(which='major',axis = 'x',direction='out', colors = Citi[2], **tkw)
#plt.xticks(x, rotation='vertical')
#host.xaxis.set_major_locator(AutoMajorLocator())
host.xaxis.set_major_locator(MultipleLocator(24))
host.tick_params('x',which='major', length=7)
#Label colours taken from plot
host.yaxis.label.set_color(p1.get_color())
par1.yaxis.label.set_color(p2.get_color())
par2.yaxis.label.set_color(p3.get_color())
# Map Title
host.set_title('Aluminium Managed Money Net Positioning as of %s'% Last_Report[0],fontsize='large')
#Colour Spines cant figure out how to do it for the host
par1.spines["right"].set_edgecolor(p2.get_color())
par2.spines["right"].set_edgecolor(p3.get_color())
###### Annotation Tests ##########
## Net Position High Box
host.annotate(f' Net Position High | {Pos_High} \n Date | {Str_Date} \n AUM | ${Pos_High_AUM[0].round(1)} Billion\n 3M Price | ${Pos_High_Price[0]}$',
xy=(Pos_High_Date, Pos_High), xycoords='data',
xytext=(0.02, .85), textcoords='axes fraction',
horizontalalignment='left',
verticalalignment='bottom',
color='white',
bbox=dict(boxstyle="round", fc= Citi[1],edgecolor='white'),
arrowprops=dict(
facecolor='black',
arrowstyle= '->'))
## Net Position Low Box
host.annotate(f' Net Position Low | {Pos_Low} \n Date | {Str_Date_Low} \n AUM | ${Pos_Low_AUM[0].round(1)} Billion\n 3M Price | ${Pos_Low_Price[0]}$',
xy=(Pos_Low_Date, Pos_Low), xycoords='data',
xytext=(0.02, .80), textcoords='axes fraction',
horizontalalignment='left',
verticalalignment='top',
color='white',
bbox=dict(boxstyle="round", fc= Citi[4],edgecolor='white'),
arrowprops=dict(
facecolor='black',
arrowstyle= '->'))
################
# Legend - a little complicated as we have to take from multiple axis
lines = [p1, p2, p3]
########## Plot text and line on chart if you want to
# host.axvline(x = DATES[52] , linestyle='dotted', color='black') ###Dotted Line when Needed
# host.text(2020.3, 10, 'Managed Money \n Aluminium')
# host.text(2020.5, 92, r'Ali',color='black')
# host.text(2020.8,15, r'some event', rotation=90)
host.legend(lines,[l.get_label() for l in lines],loc=2, fontsize=12,frameon=False)
plt.savefig('multiple_axes.png', dpi=300, bbox_inches='tight')

Avoid Overlapping of text in scatter plot with annotations

How can I avoid overlapping text in the following image? My upper scatter plot points are very closer, but it looks like, if I try anything, it doesn't change anything, Is there any work around for the same.
plt.figure(figsize=(17,15), dpi= 300)
plt.style.use('fivethirtyeight')
colors = cm.rainbow(np.linspace(0, 1, 158))
ax = plt.scatter(merged_latest['HDI'], merged_latest['electricity_access'], s=merged_latest['HDI'],c = colors, alpha=0.1, cmap = plt.get_cmap('Spectral'))
plt.ylabel('Electricity Access')
plt.xlabel('HDI')
plt.title('Electricity Access versus HDI')
old_x = old_y = 1e9 # make an impossibly large initial offset
thresh = .1 #make a distance threshold
for label, x, y in zip(list(merged_latest['Country Name']), merged_latest['HDI'], merged_latest['electricity_access']):
#calculate distance
d = ((x-old_x)**2+(y-old_y)**2)**(.5)
#if distance less than thresh then flip the arrow
flip = 1
if d < .1: flip=-2
plt.annotate(
label,
xy = (x, y), xytext = (-20*flip, 20*flip),
textcoords = 'offset points', ha = 'right', va = 'bottom',
bbox = dict(boxstyle = 'round,pad=0.5', fc = 'yellow', alpha = 0.9),
arrowprops = dict(arrowstyle = '->', connectionstyle = 'arc3,rad=0'))
old_x = x
old_y = y
plt.tight_layout()
plt.show();
from adjustText import adjust_text
texts = []
for x, y, s in zip(merged_latest['Population Growth'], merged_latest['GDP per capita'], merged_latest['Country Name']):
texts.append(plt.text(x, y, s))
adjust_text(texts, only_move={'points':'y', 'texts':'y'}, arrowprops=dict(arrowstyle="->", color='r', lw=0.5))

Changeing the labels distance from x and y ticks and choosing one color to the annotation in seaborn heat map

I am doing a correlation matrix using a seaborn heatmap, I need to :
Change the distance between the ticks and the labels of x and y.
Also, change the distance between the title and the heatmap.
Unifying the color of annotation to be either white or black.
I am using the following code:
from matplotlib import pyplot as plt
import matplotlib
import numpy as np
import seaborn as sns
#call data frame and apply correlation:
#data =
#df = pd.DataFrame(data, columns = features)
#df_small = df.iloc[:,:]#if only few parameters are needed
#correlation_mat = df_small.corr()
#Create color pallete:
def NonLinCdict(steps, hexcol_array):
cdict = {'red': (), 'green': (), 'blue': ()}
for s, hexcol in zip(steps, hexcol_array):
rgb =matplotlib.colors.hex2color(hexcol)
cdict['red'] = cdict['red'] + ((s, rgb[0], rgb[0]),)
cdict['green'] = cdict['green'] + ((s, rgb[1], rgb[1]),)
cdict['blue'] = cdict['blue'] + ((s, rgb[2], rgb[2]),)
return cdict
#https://www.december.com/html/spec/colorshades.html
hc = ['#e5e5ff', '#C7DDF2', '#8EBAE5', '#407FB7', '#00549F']#ffffff #e5e5ff
th = [0, 0.25, 0.5, 0.75, 1]
hc = hc[:0:-1] + hc # prepend a reversed copy, but without repeating the central value
cdict = NonLinCdict(np.linspace(0, 1, len(hc)), hc)
cm = matplotlib.colors.LinearSegmentedColormap('test', cdict)
corr=np.random.uniform(-1, 1, (6,6))
#plot correlation matrix:
plt.figure(figsize = (10,8))
ax=sns.heatmap(corr,center=0, linewidths=1, annot = True,cmap=cm ,square=True, vmin=-1, vmax=1,
robust=True, annot_kws={'size':16}, cbar=True,linecolor='#F6A800',xticklabels=True,
yticklabels=True)
cbar = ax.collections[0].colorbar
cbar.ax.tick_params(labelsize=10, axis='both', which='both', length=0)
cbar.set_ticks(np.linspace(-1, 1, 11))
plt.title("title", y=-1.5,fontsize = 18,)
plt.xlabel("X_parameters",fontsize = 18)
plt.ylabel("Y_paramaters",fontsize = 18)
ax.tick_params(axis='both', which='both', length=0)
ax.axhline(y=0, color='#F6A800',linewidth=4)
ax.axhline(y=corr.shape[1], color='#F6A800',linewidth=4)
ax.axvline(x=0, color='#F6A800',linewidth=4)
ax.axvline(x=corr.shape[0], color='#F6A800',linewidth=4)
#change position of lables and titles and assign colors.
plt.show()
My current output is:
Well, plt.title() has a parameter pad= to set the padding between the text of the title and the top spine of the plot (default is 6). plt.xlabel() and plt.ylabel() have a parameter labelpad= to set the distance between the axis label and the ticklabels.
sns.heatmap() has a parameter annot_kws which is a dictionary of parameters for the annotation texts. The color can be changed via sns.heatmap(..., annot_kws={'size': 16, 'color': 'black'}). Note that for readability, seaborn's default uses white for text on the dark-colored cells, and black for the light-colored cells.
from matplotlib import pyplot as plt
import matplotlib
import numpy as np
import seaborn as sns
def NonLinCdict(steps, hexcol_array):
cdict = {'red': (), 'green': (), 'blue': ()}
for s, hexcol in zip(steps, hexcol_array):
rgb = matplotlib.colors.hex2color(hexcol)
cdict['red'] = cdict['red'] + ((s, rgb[0], rgb[0]),)
cdict['green'] = cdict['green'] + ((s, rgb[1], rgb[1]),)
cdict['blue'] = cdict['blue'] + ((s, rgb[2], rgb[2]),)
return cdict
hc = ['#e5e5ff', '#C7DDF2', '#8EBAE5', '#407FB7', '#00549F'] # ffffff #e5e5ff
th = [0, 0.25, 0.5, 0.75, 1]
hc = hc[:0:-1] + hc # prepend a reversed copy, but without repeating the central value
cdict = NonLinCdict(np.linspace(0, 1, len(hc)), hc)
cm = matplotlib.colors.LinearSegmentedColormap('test', cdict)
corr = np.random.uniform(-1, 1, (6, 6))
# plot correlation matrix:
plt.figure(figsize=(10, 8))
ax = sns.heatmap(corr, center=0, linewidths=1, annot=True, cmap=cm, square=True, vmin=-1, vmax=1,
robust=True, annot_kws={'size': 16, 'color': 'black'}, cbar=True, linecolor='#F6A800',
xticklabels=True, yticklabels=True)
cbar = ax.collections[0].colorbar
cbar.ax.tick_params(labelsize=10, axis='both', which='both', length=0)
cbar.set_ticks(np.linspace(-1, 1, 11))
plt.title("title", y=-1.5, fontsize=18, pad=15)
plt.xlabel("X_parameters", fontsize=18, labelpad=15)
plt.ylabel("Y_paramaters", fontsize=18, labelpad=15)
ax.tick_params(axis='both', which='both', length=0)
ax.axhline(y=0, color='#F6A800', linewidth=4)
ax.axhline(y=corr.shape[1], color='#F6A800', linewidth=4)
ax.axvline(x=0, color='#F6A800', linewidth=4)
ax.axvline(x=corr.shape[0], color='#F6A800', linewidth=4)
plt.show()

How to show exact decimals in horizontal stacked bar chart

I have found a code for creating stacked bar charts on https://matplotlib.org/3.1.1/gallery/lines_bars_and_markers/horizontal_barchart_distribution.html.
Problem is that decimal places are not shown in the graph. In the code i've found there are no decimals in the list, but if you change them to decimals there are not shown (see code below).
If tried to change str to float in ax.text(x, y, str(int(c)) line. That actually shows decimals, but all of them are 0.
many thanks.
import numpy as np
import matplotlib.pyplot as plt
category_names = ['Strongly disagree', 'Disagree','Neither agree nor
agree', 'agree', 'Strongly agree']
results = {
'Question 1': [(10), (14.99), (17.01), (32), (26)]
}
def survey(results, category_names):
labels = list(results.keys())
data = np.array(list(results.values()))
data_cum = data.cumsum(axis=1)
category_colors = plt.get_cmap('RdYlGn')(
np.linspace(0.15, 0.85, data.shape[1]))
fig, ax = plt.subplots(figsize=(9.2, 5))
ax.invert_yaxis()
ax.xaxis.set_visible(False)
ax.set_xlim(0, np.sum(data, axis=1).max())
for i, (colname, color) in enumerate(zip(category_names, category_colors)):
widths = data[:, i]
starts = data_cum[:, i] - widths
ax.barh(labels, widths, left=starts, height=0.5,
label=colname, color=color)
xcenters = starts + widths / 2
r, g, b, _ = color
text_color = 'white' if r * g * b < 0.5 else 'darkgrey'
for y, (x, c) in enumerate(zip(xcenters, widths)):
ax.text(x, y, str(int(c)), ha='center', va='center',
color=text_color)
ax.legend(ncol=len(category_names), bbox_to_anchor=(0, 1),
loc='lower left', fontsize='small')
return fig, ax
survey(results, category_names)
When you write str(int(c)), you first convert the number to integer, thereby dropping the decimals, and then convert that into a string. To get the desired output, either use str(c), or if you want to be able to control the format of the number better (such as choosing the number of digits after the decimal point) replace str(int(c)) by '{:.2f}'.format(c)
category_names = ['Strongly disagree', 'Disagree','Neither agree nor agree', 'agree', 'Strongly agree']
results = {
'Question 1': [(10), (14.99), (17.01), (32), (26)]
}
def survey(results, category_names):
labels = list(results.keys())
data = np.array(list(results.values()))
data_cum = data.cumsum(axis=1)
category_colors = plt.get_cmap('RdYlGn')(
np.linspace(0.15, 0.85, data.shape[1]))
fig, ax = plt.subplots(figsize=(9.2, 5))
ax.invert_yaxis()
ax.xaxis.set_visible(False)
ax.set_xlim(0, np.sum(data, axis=1).max())
for i, (colname, color) in enumerate(zip(category_names, category_colors)):
widths = data[:, i]
starts = data_cum[:, i] - widths
ax.barh(labels, widths, left=starts, height=0.5,
label=colname, color=color)
xcenters = starts + widths / 2
r, g, b, _ = color
text_color = 'white' if r * g * b < 0.5 else 'darkgrey'
for y, (x, c) in enumerate(zip(xcenters, widths)):
ax.text(x, y, '{:.2f}'.format(c), ha='center', va='center',
color=text_color)
ax.legend(ncol=len(category_names), bbox_to_anchor=(0, 1),
loc='lower left', fontsize='small')
return fig, ax
survey(results, category_names)

Categories