Seaborn: distplot
from matplotlib import pyplot as plt
import seaborn as sns
cars = sns.load_dataset('mpg').dropna()
cars.shape
(392, 9)
cars.head()
mpg cylinders displacement horsepower weight acceleration \
0 18.0 8 307.0 130.0 3504 12.0
1 15.0 8 350.0 165.0 3693 11.5
2 18.0 8 318.0 150.0 3436 11.0
3 16.0 8 304.0 150.0 3433 12.0
4 17.0 8 302.0 140.0 3449 10.5
model_year origin name
0 70 usa chevrolet chevelle malibu
1 70 usa buick skylark 320
2 70 usa plymouth satellite
3 70 usa amc rebel sst
4 70 usa ford torino
sns.set_style('white')
Intro Visuals
hp = cars.horsepower.sample(100, random_state=100)
plt.rc('xtick', labelsize=14)
plt.rc('ytick', labelsize=14)
plt.figure(figsize=(8,5))
sns.distplot(hp, rug=True, kde=False, hist=False, rug_kws={'lw': 2})
sns.despine(left=True)
plt.xlim(0, 280)
plt.xlabel('')
plt.yticks([])
plt.tight_layout();
plt.figure(figsize=(8,5))
sns.distplot(hp, kde=False, hist_kws={'alpha': 1})
sns.despine(left=True)
plt.xlim(0, 280)
plt.xlabel('')
plt.yticks([])
plt.tight_layout();
plt.figure(figsize=(8,5))
sns.distplot(hp, hist=False, kde_kws={'lw': 5})
sns.despine(left=True)
plt.xlim(0, 280)
plt.xlabel('')
plt.yticks([])
plt.tight_layout();
plt.figure(figsize=(8,5))
sns.distplot(hp, rug=True, kde_kws={"lw": 5}, rug_kws={'lw': 2} )
sns.despine(left=True)
plt.xlim(0, 280)
plt.xlabel('')
plt.yticks([])
plt.tight_layout();
Basics
sns.distplot(cars.weight);
sns.distplot(cars.weight, kde=False);
sns.distplot(cars.weight, hist=False, rug=True);
Options
Bins
sns.distplot(cars.weight, bins=5);
sns.distplot(cars.weight, bins=[1500, 2000, 2500, 3500, 5200]);
Fit
from scipy.stats import skewnorm
sns.distplot(cars.weight, fit=skewnorm, kde=False);
Kwargs
hist_kws
sns.distplot(cars.weight, kde=False, hist_kws={'alpha': 1});
sns.distplot(cars.weight, kde=False,
hist_kws={'alpha': 1, 'histtype': 'stepfilled'});
kde_kws
sns.distplot(cars.weight, kde_kws={'lw': 5});
sns.kdeplot(cars.weight, lw=5);
sns.distplot(cars.weight, kde_kws={'bw': 1000});
rug_kws
sns.distplot(cars.weight, rug=True, rug_kws={'height': 0.15});
sns.distplot(cars.weight, rug=True,
rug_kws={'height': 0.15, 'color': 'xkcd:navy'});
Styled Distplots
sns.set_style('white')
plt.figure(figsize=(8,5))
sns.distplot(cars.weight,
bins=20,
kde_kws={"lw": 8, 'color':'xkcd:bluish green'},
hist_kws={'alpha': 0.3, 'color':'gray'}
)
sns.despine(left=True)
plt.xlabel('Weight (lbs)', fontsize=16, labelpad=15)
plt.yticks([])
plt.tight_layout()
plt.savefig('03dist_styleWeight_greenKDE.svg');
sns.set_style('white')
plt.figure(figsize=(8,5))
sns.distplot(cars.acceleration.sample(100, random_state=100),
rug=True,
kde=False,
hist_kws={'lw': 5,
'linestyle': (0, (5, 1)),
'alpha': 1,
'histtype':'step',
'color':'xkcd:dark magenta'},
rug_kws={'lw': 3,
'color':'xkcd:dark magenta'}
)
sns.distplot(cars.acceleration.sample(100, random_state=100),
kde=False,
hist_kws={'alpha': 0.1,
'histtype':'stepfilled',
'color':'xkcd:dark magenta'},
)
sns.despine(left=True)
plt.xlabel('')
plt.xticks([])
plt.yticks([])
plt.tight_layout()
plt.savefig('03dist_styleAcc_pink.svg');
import random
random.seed(11)
data = [random.gauss(2, 0.75) for _ in range(50)]
data.extend([random.gauss(-2, 0.75) for _ in range(50)])
sns.set_style('whitegrid')
plt.figure(figsize=(8,5))
sns.distplot(data,
rug=True,
hist=False,
color='green',
kde_kws={'lw':4, 'shade':True, 'bw':.5},
rug_kws={'lw':2}
)
sns.despine()
plt.xlabel('')
plt.tight_layout()
#plt.savefig('03dist_styleRand_greenKDErug.svg');
sns.set_style('darkgrid')
data = [random.gauss(2, 0.75) for _ in range(100)]
data.extend([random.gauss(-2, 0.75) for _ in range(50)])
plt.figure(figsize=(8,5))
sns.distplot(data,
hist_kws={'histtype': 'stepfilled',
'alpha':0.7, 'color':'xkcd:carolina blue'},
kde_kws={'lw':8,'color':'xkcd:navy'}
)
sns.despine()
plt.xlabel('')
plt.tight_layout()
#plt.savefig('03dist_styleRand_carolina.svg');
