How to plot based upon unique column values? - python-3.x

Question

I'm a beginner learning to use python to do data visualizations.
I found a really cool data set by the UN it is formatted like this:
Afghanistan 1975 2127
Afghanistan 1985 3509
Afghanistan 1995 1243
Afghanistan 2005 1327
Albania 1975 4595
Albania 1985 7880
Albania 1995 2087
Albania 2005 4254
etc...
Up until now, I've been parsing out individual countries with statements like this:
china = data[data.area == 'China']
This is fine for picking individual countries but now, I want to plot all of them. How could I go about that?
So far I've tried this but couldn't figure out how to make it work:
old_value = data.iloc[0]
for i in len(data):
if data.iloc[i].area == old_value:
# add to current set
else:
# create new set
Any help would be much appreciated!

Answer 1

Given your data
Setup imports and dataframe
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
# plot parameters
plt.style.use('seaborn')
plt.rcParams['figure.figsize'] = (16.0, 10.0)
data = {'country': ['Afghanistan', 'Afghanistan', 'Afghanistan', 'Afghanistan', 'Albania', 'Albania', 'Albania', 'Albania'],
'year': [1975, 1985, 1995, 2005, 1975, 1985, 1995, 2005],
'value': [2127, 3509, 1243, 1327, 4595, 7880, 2087, 4254]}
df = pd.DataFrame(data)
country year value
0 Afghanistan 1975 2127
1 Afghanistan 1985 3509
2 Afghanistan 1995 1243
3 Afghanistan 2005 1327
4 Albania 1975 4595
5 Albania 1985 7880
6 Albania 1995 2087
7 Albania 2005 4254
Use seaborn.barplot with the hue parameter
p = sns.barplot(x='year', y='value', hue='country', data=df)
Horizontally
p = sns.barplot(x='value', y='year', hue='country', data=df, orient='h')
A separate plot for each country
Using plt.subplot(1, 2, i) the rows times the columns should equal the number of unique countries or +1 if there are an odd number.
max_value = df.value.max() + 100 # + 100 to add padding at the top of the plot; 100 is an arbitrary value and can be removed
for i, country in enumerate(df.country.unique(), 1): # iterate through each unique country
data = df[df.country == country] # filter by country
plt.subplot(1, 2, i) # rows, columns, i: plot index beginning at 1
sns.barplot(x='year', y='value', data=data)
plt.ylim(0, max_value) # set y-lim with max of the value column; makes it easier to compare countries
plt.title(country)