Pandas

Question 1

series

Answer 1

array with index

Question 2

Example: create a series

Answer 2

obj = Series([4, 7, -5, 3], index = [‘d’, ‘b’, ‘a’, ‘c’])

Question 3

find obj value

Answer 3

obj.values

Question 4

find obj index

Answer 4

obj.index

Question 5

Find value for index ‘a’ and index ‘c’,’a’,’d’

Answer 5

1. obj[‘a’]

2. obj[[‘c’, ‘a’, ‘d’]]

Question 6

Series vs. dict

Answer 6

Series is a fixed-length, ordered dict

Question 7

Check if index ‘b’ exists in obj

Answer 7

‘b’ in obj

Question 8

Create series from dict sdata

Answer 8

obj = Series (sdata)

Question 9

Check if the value exits (or not exits) for the series obj

Answer 9

1. pd.isnull(obj)

2. pd.notnull(obj)

Question 10

Create name attributes for Series object obj and its index

Answer 10

1. obj.name = ‘population’

2. obj.index.name = ‘state’

Question 11

Most common ways to construct DataFrame

Answer 11

1. from a dict of equal-length lists e.g. if data is a dict object, then DataFrame object frame = DataFrame(data)
2. NumPy arrays, e.g. if data is a NumPy array, DataFrame(data, columns=[‘year’, ‘state’, ‘pop’]

Question 12

How to retrieve a column from DataFrame as a Series

Answer 12

1. dict-like notation: frame[‘state’]

2. by attribute. frame.year

Question 13

retrieve a row from DAta Frame object frame with index ‘three’

Answer 13

frame.ix[‘three’]

Question 14

delete a column ‘eastern’ from DataFrame object frame

Answer 14

del frame[‘eastern’]

Question 15

Is the index objects mutable?

Answer 15

No, so Index objects c an be safely shared among data structures

Question 16

create an Index object

Answer 16

index =- pd.Index(np.arrange(3))

Question 17

Class: Int64Index

Answer 17

Specialized Index for integer values

Question 18

Class: MultiIndex

Answer 18

“Hierarchical” index object representing multiple levels of indexing on a single axis

Question 19

Class: DatetimeIndex

Answer 19

Stores nanosecond timestamps (represented using NumPy’s datatime64 dtype)

Question 20

Class: PeriodIndex

Answer 20

Specialized Index for Period data (timespans)

Question 21

Index methods

Answer 21

append, diff, intersection, union, isin, delete, drop, insert, is_monotonic, is_unique, unique

Question 22

reindex

Answer 22

Crate a new object with the data conformed to a new index. e.g. obj2 = obj.reindex([‘a’, ‘b’, ‘c’, ‘d’, ‘e’], fill_value=0)

Question 23

forward fill when reindexing

Answer 23

ffill or pad; obj3.reindex(range(6), method = ‘ffill’)

Question 24

backward fill when reindexing

Answer 24

bfill or backfill

Question 25

Drop entry (1) ‘c’ (2) ‘d’ and ‘c’

Answer 25

obj. drop(‘c’)

obj. drop([‘d’, ‘c’])

Question 26

special indexing field ix

Answer 26

Select a subset of the rows and columns from a DataFrame with NumPy like notation plus axis labels

Question 27

Hierarchical indexing

Answer 27

Enable you to have multiple index levels on an axis