Pandas

Question 1

Pandas

Answer 1

python library
easy to load and manipulate data
integrates with loads of analysis and visualization libraries

Question 2

dataframes

Answer 2

specific structure for two-dimensional data

pd.Dataframe()

Question 3

What method do you use to sort by a series in pandas?

Answer 3

sort_values(by = ‘series’ (this can be a list so you can sort by more than one column),)

Question 4

which method to list the count of unique values in a series

Answer 4

value_counts()

Question 5

List the breed of the cat with the highest price

Answer 5

cat_df[‘breed’,’price’].max()

Question 6

List info on cats with breed tabby, calico, or bengal

Answer 6

cat_df.loc[(cat_df[‘breed’] == ‘calico’) | (cat_df[‘breed’] == ‘tabby’) | (cat_df[‘breed’] == ‘bengal’)]

Question 7

In a dataset with car company’s and price find each company’s highest priced car

Answer 7

price = new[‘company’, ‘price’].max()

Question 8

how to do a join on pandas ex) on column ‘cat’

Answer 8

pd.merge(df2, df2, on = ‘cat’)

Question 9

how to create a dataframe from a dictionary

Answer 9

pd.DataFrame.From_dict()

Question 10

method to get rid of duplicates

Answer 10

pd.drop_duplicates

Question 11

All the unique items

Answer 11

df.unique() or nunique for number of unique items

Question 12

combine datasets

Answer 12

make a list of dataframes then df.concat()

Question 13

in a cat dataframe how to get all info where cat breed = ‘tabby’ or ‘calico

Answer 13

df.loc[(df[breed] == tabby) | (df[breed] == calico)]

if only one condition no need for parentheses

Question 14

add a row to dataframe

Answer 14

df.append(dataframe or list)

Question 15

iterate through a dataframe

Answer 15

df.iterrows() will return index, row of dataframe
ex)
for i, row in df.itterrows():
wins = row[w]
games = row[g]
win perc = win/games_played
list.append(winperc)

df[winspercentaget) = list

Question 16

grab columns as numpy arrays

Answer 16

use .values
runn_diffs_np = baseball_df[run].values - baseball_df[‘ra].values
will give you a numpy array that you can append to the dataframe

Question 17

identify na values when creating dataframe

Answer 17

na_values argument in read_csv

Question 18

replace na values

Answer 18

df.fillna()
df.fillna(0) - will fill with zeroes
values = {“A”: 0, “B”: 1, “C”: 2, “D”: 3}
df.fillna(value=values) - will fill columns with appropriate na values
df.fillna(value=values, limit=1) - will only replace first nan value