1import pandas as pd # import pandas package
2import numpy as np

1dataquest_aio = '<https://raw.githubusercontent.com/dinhanhthi/dataquest-aio/master/step-2-data-analysis-and-visualization/>'
2dataset_url = dataquest_aio + 'course-4-data-cleaning-and-analysis/data/World_Happiness_2015.csv'
3df = pd.read_csv(dataset_url) # read the data set
4df.head()

1df.groupby('Region').get_group('Western Europe') # returns a df

1df.groupby('Region')['Happiness Score'].mean()
2# other methods: size, max, min, count

1def max_min(group):
2  return group.max() - group.min()
3
4df.groupby(['Country', 'Region']).agg([np.mean, np.max, max_min]).head()

1df.groupby(['Country', 'Region']).agg({
2    'Happiness Rank': max_min,
3    'Happiness Score': ['min', 'max'],
4    'Standard Error': 'count'
5}).head(3)

1orders.groupby('shoes').price.apply(lambda x: np.min(x, 25)).reset_index()

1df.pivot_table(values=['GDP', 'City'], index='Region') # returns df

1df.pivot_table(['GDP', 'City'], 'Region', aggfunc=[np.mean, np.max], margins=True)
2# margins shows the "All" row

1df = pd.DataFrame({'foo': ['one', 'one', 'one', 'two', 'two',
2                           'two'],
3                   'bar': ['A', 'B', 'C', 'A', 'B', 'C'],
4                   'baz': [1, 2, 3, 4, 5, 6],
5                   'zoo': ['x', 'y', 'z', 'q', 'w', 't']})
6
7pivot_1 = df.pivot(index='foo', columns='bar', values='baz')
8pivot_2 = df.pivot(index='foo', columns='bar')['baz']
9pivot_3 = df.pivot(index='foo', columns='bar', values=['baz', 'zoo'])
10
11display_side_by_side(df, pivot_1, pivot_2, pivot_3)

1df = pd.DataFrame({'A': {0: 'a', 1: 'b', 2: 'c'},
2                   'B': {0: 1, 1: 3, 2: 5},
3                   'C': {0: 2, 1: 4, 2: 6}})
4
5df1 = pd.melt(df, id_vars=['A'], value_vars=['B'])
6df2 = pd.melt(df, id_vars=['A'], value_vars=['B', 'C'])
7
8display_side_by_side(df, df1, df2)