Difference between [] and [[]] in Python

user_14036406 0
Viewed 2533

I am taking a course on Coursera titled "Data Analysis with Python". I am new to Python. I have some experience with C and MATLAB. That's why I haven't faced that much problem except for one thing.

At first please take a look at the following code.

import pandas as pd

#path of data
path = 'https://cf-courses-data.s3.us.cloud-object-storage.appdomain.cloud/IBMDeveloperSkillsNetwork-DA0101EN-SkillsNetwork/labs/Data%20files/automobileEDA.csv'
df = pd.read_csv(path)

from sklearn.linear_model import LinearRegression

lm = LinearRegression()

X = df[['highway-mpg']] #'highway-mpg' is a column in the dataframe
Y = df['price'] #'price' is a column in the dataframe

lm.fit(X,Y)

Here while defining X they have used double square brackets [[]] and in the case of Y used []. I know that square brackets are used for listing in Python and nested listing is possible. But it doesn't seem to be a nested list. Can you please differentiate between them?

pandas python
5 Answers

One does slice the dataframe and return a sub-dataframe, the other select exactly one column and gives it as a Series:

  • If you pass a list of string to the brackets [['a', 'c']], you'll obtain DataFrame which contains only the column you asked

  • If you pass a string to the brackets ['a'], you'll obtain only a Series which correponds to the name you give

Example

df = pd.DataFrame([
    {'a': 1, 'b': 2, 'c': 4},
    {'a': 3, 'b': 4, 'c': 6},
])

print(df['a'])
# That is a Series
0    1
1    3
Name: a, dtype: int64
# -------------------------------------------
print(df[['a']])
# That is a DataFrame
   a
0  1
1  3
# -------------------------------------------
print(df[['a', 'c']])
# That is a DataFrame
   a  c
0  1  4
1  3  6
user_72126860

when you want to use or return a single column of the data frame i.e. returns a Series then use df['xyz'] in case you want to return or use multiple columns you can use df[['xyz','abc','ijk']] which returns another data frame with specified columns.

Note

df[['xyz']] this is allowed, but df['xyz','abc','ijk'] this is not allowed.

df1=pd.DataFrame({"Col":[1,2,3],"Row":[2,3,4],"value":[6,5,7],"ree":[0,0,0]})
df1


  Col Row  value ree
0   1   2   6   0
1   2   3   5   0
2   3   4   7   0

df1['Col']
0    1
1    2     # Series object
2    3
Name: Col, dtype: int64

df1[['Col']]
    Col
0   1
1   2      # DataFrame object
2   3

df1[["Col","Row","value"]]
    Col Row value
0   1   2   6
1   2   3   5         # DataFrame object
2   3   4   7

in your case X['h___...'] can also be used as there is only one column

user_124432600

[[]] is generally used to pass multiple columns to retrieve from a dataframe. Another key distinction is it returns a DataFrame

[] is generally used to pass a single column to retrieve from a dataframe. Another key distinction is it returns a Series

This is one of the key differences when indexing a dataframe. Both perform the same intended operation with different return types

user_91089120

In Machine Learning and Data Analysis, usually X denotes independent variables and y denotes dependent variable. The goal is to predict single dependent variable y using two or many independent variables X. And, In Pandas [[]] is used to get variable number of columns. That is why [[]] used for X and [] for y

user_146864060

[[colname1,colname2,colnamen]] so you are selecting multiple columns from the dataframe whereas ["target_name"] you only need to specify the name of the column since it's just one.

user_129592410
Related