Python - Pandas Library

Question 1

Syntax
Creating DataFrames

Create a DataFrame using dict format, specifying values for columns

Answer 1

df = pd.DataFrame( { “a” : [4 ,5, 6],
“b” : [7, 8, 9],
“c” : [10, 11, 12] },
index = [1, 2, 3] )

Question 2

Syntax

Creating DataFrames

Create a DataFrame using list format, specifying values for each row

Answer 2

df = pd.DataFrame( [[4, 7, 10], [5, 8, 11], [6, 9, 12] ],

index=[1, 2, 3],

columns=[‘a’, ‘b’, ‘c’] )

Question 3

Method Chaining

Most Pandas methods returns a DF so another method can be applied to the result.

Melt a DF, rename the columns, and then query it.

Answer 3

df = ( pd.melt(df)

.rename( columns={ ‘variable’ : ‘var’, ‘value’ : ‘val’} )

.query(‘val >= 200’)

)

Question 4

Reshaping Data

Gather the columns into rows

Answer 4

pd.melt (df)

Question 5

Reshaping Data

Spread rows into columns

Answer 5

df.pivot ( columns=’var’,

values=’val’

)

Question 6

Reshaping Data

Append rows of df2 to df1

Answer 6

pd.concat ( [df1, df2] )

Question 7

Reshaping Data

Append columns of df2 to df1

Answer 7

pd.concat ( [df1, df2],

axis=1

)

Question 8

Data Structures

What’s a Series and how do you define it?

Answer 8

A Series is basically an indexed list.

s = pd.Series ( [3, -5, 7, 4],

index=[‘a’, ‘b’, ‘c’, ‘d’]

)

Question 9

Data Structures

What’s a DataFrame and how do you define it?

Answer 9

A DataFrame is basically a bunch of Series (or columns) that are concatenated together and indexed.

data = { ‘Country’: [‘Belgium’, ‘India’, ‘Brazil’],

‘Capital’: [‘Brussels’, ‘New Delhi’, ‘Brasília’],

‘Population’: [11190846, 1303171035, 207847528]

}

df = pd.DataFrame ( data,

columns=[‘Country’, ‘Capital’, ‘Population’]

)

Question 10

Selection

How do you get one column of values?

Answer 10

You access it like a dictionary (by referencing the key or in this case the column name)

col_of_values = df [‘col_name’]

Question 11

Selection

How do you get a subset of the DataFrame, or get multiple columns?

Answer 11

Access it just like list indexing / slicing.

df [start_col# : stop_col#]

Question 12

Selecting

Select value(s) in a row with a named index.

Answer 12

df.loc [name_of index or row]

Question 13

Selecting

Select value(s) in a row with a named index with a certain condition, and only those specific columns.

Answer 13

df.loc [df [ ‘a’] > 10,

[‘a’, ‘c’] ]

Question 14

Selecting

Select value(s) in a column with a numbered / default index.

Answer 14

df.iloc [num_of_indexed_row]

Question 15

Subset Observations / Rows

Extract rows that meet a logical condition

Answer 15

df [df.Length > 7]

Question 16

Subset Observations / Rows

Remove duplicate rows

Answer 16

df.drop_duplicates()

Question 17

Subset Observation / Rows

Select first n rows

Answer 17

df.head(n)

Question 18

Subset Observation / Rows​

Select last n rows

Answer 18

df.tail(n)

Question 19

Subset Observation / Rows​

Select rows by default indexed position

Answer 19

df.iloc [10:20]

Question 20

Subset Variables / Columns

Select multiple columns with specific names.

Answer 20

df [[ ‘width’, ‘length’, ‘species’] ]

*notice you have to input a list into the dataframe’s reference

Question 21

Subset Variables / Columns

Select single column with specific name.

Answer 21

df [‘width’]

or

df.width

Question 22

Parameter that can be inserted in most DataFrame methods to modify the original DataFrame in place.

Answer 22

inplace = ‘True’

Question 23

Sort

Order the rows by values of a column (low to high)

Answer 23

df.sort_values ( ‘mpg’ )

Question 24

Sort

Order rows by values of a column (from high to low)

Answer 24

df.sort_values ( ‘mpg’, ascending=False )

Question 25

Rename

Rename the columns of a DataFrame

Answer 25

df.rename( columns = { ‘y’:’year’ } )

Question 26

Drop

Drop certain columns from a DataFrame

Answer 26

df.drop ( columns=[‘Length’, ‘Height’] )

Question 27

Handling Missing Data

Drop rows with any column having NA/null data

Answer 27

df.dropna()

Question 28

Handling Missing Data

Replace all NA/null data with a value.

Answer 28

df.fillna( value )

Question 29

Handling Missing Data

Replace values with others

Answer 29

df.replace ( ‘a’, ‘f’ )

Question 30

Duplicate Data

Return unique values

Answer 30

df.unique()

Question 31

Duplicate Data

Drop duplicates

Answer 31

df.dropt_duplicates()

Question 32

Summarize Data

Count the number of rows with non-null values of each object

Answer 32

df [‘w’] .counts()

…or if you want to get rid of duplicates too…

df [‘w’] .value_counts()

Question 33

Summarize Data

Give # rows in a DataFrame

Answer 33

len(DF)

Question 34

Summarize Data

Give the # of rows and columns of a DataFrame

Answer 34

df.shape()

Question 35

Summarize Data

Describe the DataFrame columns

Answer 35

df.columns()

Question 36

Summarize Data

Describe the basic DataFrame information

Answer 36

df.info()

Question 37

Summarize Data

Describe the basic descriptive statistics for each column (or GroupBy)

Answer 37

df.describe()

Question 38

Summarize Data

Sum the values of a column

Answer 38

df.sum()

Question 39

Summarize Data

Give the median and mean of a column.

Answer 39

df. median()
df. mean()

Question 40

Summarize Data

Give the min, max, var, std of a column(s).

Answer 40

df. min()
df. max()
df. var()
df. std()

Question 41

Apply Function

Apply a function to each value (all rows) of a column

*Note this is vectorizes, so it’s efficient + fast

Answer 41

function = lambda x: x*2

df.apply( function )