Lecture 6 – Data 100, Fall 2022

by Josh Hug

adapted from material by Ani Adhikari, Suraj Rampure, and Fernando Pérez.

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

births = pd.read_csv('baby.csv')

plt.rcParams["hist.bins"]

10

births.head()

	Birth Weight	Gestational Days	Maternal Age	Maternal Height	Maternal Pregnancy Weight	Maternal Smoker
0	120	284	27	62	100	False
1	113	282	33	64	135	False
2	128	279	28	64	115	True
3	108	282	23	67	125	True
4	136	286	25	62	93	False

births.shape

(1174, 6)

Bar Plots

We often use bar plots to display distributions of a categorical variable:

births['Maternal Smoker'].value_counts()

False    715
True     459
Name: Maternal Smoker, dtype: int64

This would have been the Data 8 code to do something similar:

from datascience import Table
t = Table.from_df(births['Maternal Smoker'].value_counts().reset_index())
t.barh("index", "Maternal Smoker")

births['Maternal Smoker'].value_counts().plot(kind = 'bar');

ms = births['Maternal Smoker'].value_counts();
plt.bar(ms.index, ms);

Note: putting a semicolon after a plot call hides all of the unnecessary text that comes after it (the <matplotlib.axes_....>).

sns.countplot(data = births, x = 'Maternal Smoker');

import plotly.express as px
px.histogram(births, x = 'Maternal Smoker', color = 'Maternal Smoker')

sns.countplot(data = births, x = 'Maternal Pregnancy Weight');

sns.histplot(data = births, x = 'Maternal Pregnancy Weight');

px.histogram(births, x = 'Maternal Pregnancy Weight')

sns.histplot(data = births, x = 'Maternal Pregnancy Weight', bins = 20);
sns.rugplot(data = births, x = 'Maternal Pregnancy Weight', color = "red");

sns.histplot(data = births, x = 'Maternal Pregnancy Weight', kde = True);
sns.rugplot(data = births, x = 'Maternal Pregnancy Weight', color = "red");

Box Plots

plt.figure(figsize = (3, 6))
sns.boxplot(y = "Birth Weight", data = births);

bweights = births["Birth Weight"]
q1 = np.percentile(bweights, 25)
q2 = np.percentile(bweights, 50)
q3 = np.percentile(bweights, 75)
iqr = q3 - q1
whisk1 = q1 - 1.5*iqr
whisk2 = q3 + 1.5*iqr

whisk1, q1, q2, q3, whisk2

(73.5, 108.0, 120.0, 131.0, 165.5)

Violin Plots

plt.figure(figsize = (3, 6))
sns.violinplot(y=births["Birth Weight"]);

Side by side box plots and violin plots

plt.figure(figsize=(5, 8))
sns.boxplot(data=births, x = 'Maternal Smoker', y = 'Birth Weight');

plt.figure(figsize=(5, 8))
sns.violinplot(data=births, x = 'Maternal Smoker', y = 'Birth Weight');

Scatter plots

births.head()

	Birth Weight	Gestational Days	Maternal Age	Maternal Height	Maternal Pregnancy Weight	Maternal Smoker
0	120	284	27	62	100	False
1	113	282	33	64	135	False
2	128	279	28	64	115	True
3	108	282	23	67	125	True
4	136	286	25	62	93	False

plt.scatter(births['Maternal Height'], births['Birth Weight']);
plt.xlabel('Maternal Height')
plt.ylabel('Birth Weight');

plt.scatter(data=births, x='Maternal Height', y='Birth Weight');
plt.xlabel('Maternal Height')
plt.ylabel('Birth Weight');

sns.scatterplot(data = births, x = 'Maternal Height', y = 'Birth Weight', hue = 'Maternal Smoker');

births["Maternal Height (jittered)"] = births["Maternal Height"] + np.random.uniform(-0.2, 0.2, len(births))
fig = sns.scatterplot(data = births, x = 'Maternal Height (jittered)', y = 'Birth Weight', hue = 'Maternal Smoker');

sns.lmplot(data = births, x = 'Maternal Height', y = 'Birth Weight', ci=False, hue='Maternal Smoker');

sns.jointplot(data = births, x = 'Maternal Height', y = 'Birth Weight');

sns.jointplot(data = births, x = 'Maternal Height', y = 'Birth Weight', hue='Maternal Smoker');

Hex plots and contour plots

sns.jointplot(data = births, x = 'Maternal Height', y = 'Birth Weight', kind='hex');

sns.jointplot(data = births, x = 'Maternal Height', y = 'Birth Weight', kind='kde', fill=True);

sns.jointplot(data = births, x = 'Maternal Height', y = 'Birth Weight', kind='kde', hue='Maternal Smoker');

Bonus

Calling the DataFrame .plot() method results in weird things!

births.plot();