How to read a CSV file with pandas

Use pd.read_csv("file.csv"). It reads the file and returns a DataFrame, with column names taken from the first line.

Load a CSV and look at it

Save this as people.csv:

name,city,age,score
Ada,London,36,91.5
Ben,Paris,29,78.0
Cara,Berlin,41,85.25
Dev,Delhi,33,88.0

Then, in a script in the same folder:

import pandas as pd

df = pd.read_csv("people.csv")
print(df.head(2))
print(df.shape)
df.info()
print(pd.read_csv("people.csv", usecols=["name", "score"]))

Output:

  name    city  age  score
0  Ada  London   36   91.5
1  Ben   Paris   29   78.0
(4, 4)
<class 'pandas.DataFrame'>
RangeIndex: 4 entries, 0 to 3
Data columns (total 4 columns):
 #   Column  Non-Null Count  Dtype  
---  ------  --------------  -----  
 0   name    4 non-null      str    
 1   city    4 non-null      str    
 2   age     4 non-null      int64  
 3   score   4 non-null      float64
dtypes: float64(1), int64(1), str(2)
memory usage: 295.0 bytes
   name  score
0   Ada  91.50
1   Ben  78.00
2  Cara  85.25
3   Dev  88.00

What each line does

• head(2) shows the first two rows. With no number, it shows five.
• shape is (rows, columns), so (4, 4) means four rows and four columns.
• info() lists each column's type and how many values are not missing. It prints by itself, so you don't wrap it in print(). The memory figure may differ on your computer.
• usecols= loads only the columns you list. It's handy for wide files.

Check the types every time. Here age is int64 and score is float64, because pandas works them out from the values.

Common mistake

FileNotFoundError. The path is relative to the folder you run Python from, not the folder the script is in. If the file is in data/, write pd.read_csv("data/people.csv").

Wrong separator. If everything lands in one column, the file probably uses semicolons. Add sep=";".

Call pd.read_csv(path), then check head(), shape and info() before you do anything else with the data.