Pandas DataFrame useful functions

Let's see how to create Pandas DataFrame from the Python collection data or other data frames.

1. Create from column values.

From dict by column:

import pandas as pd
df = pd.DataFrame({'col1': [1,2], 'col2': [3,4]})
df

Out:
   col1  col2
0     1     3
1     2     4

2. Create from rows.

From rows list where each row is list of values:

d = pd.DataFrame(
    [['a', 'b'], ['c', 'd']],
    columns=['col1', 'col2']
)
d

Out: 
  col1 col2
0    a    b
1    c    d

3. Combine 2 data frames with same columns into one.

Append rows of 2 data frames:

pd.concat([df, d])

Out: 
  col1 col2
0    1    3
1    2    4
0    a    b
1    c    d

4. Join data frames like SQL tables.

Merge two data frames by the column values:

products = pd.DataFrame(
    [['Pencil', 0.65], ['Notebook', 2.99]],
    columns=['Name', 'Price']
)
products

Out: 
       Name  Price
0    Pencil   0.65
1  Notebook   2.99

orders = pd.DataFrame([
    ['Pencil', 5],
    ['Pencil', 10],
    ['Notebook', 2],
    ['Pencil', 2],
    ['Notebook', 3]
  ],
  columns=['Name', 'Quantity']
)
orders
Out: 
       Name  Quantity
0    Pencil         5
1    Pencil        10
2  Notebook         2
3    Pencil         2
4  Notebook         3

df = pd.merge(orders, products, on='Name')
df

Out: 
       Name  Quantity  Price
0    Pencil         5   0.65
1    Pencil        10   0.65
2    Pencil         2   0.65
3  Notebook         2   2.99
4  Notebook         3   2.99

5. Make aggregation.

Aggregate by Name column calculating the Amount for each name:

df['Amount'] = df['Quantity'] * df['Price']  # add amount column
df.groupby(['Name']).aggregate({'Amount': 'sum'})

Out: 
          Amount
Name            
Notebook   14.95
Pencil     11.05