在熊猫数据框中创建新列

Question

I'm extremely new to python and have been searching google and stackoverflow to solve this issue which I am sure is simply a syntax problem.

I have a data frame with several columns.

import pandas as pd
df = pd.read_csv("C:/path/file.csv")

My csv has 5 columns and ~ 100k rows I simply want a substring of the first 2 digits of column 5.

I've tried:

df.assign(new = lambda x: x.column5[0:2],)

This creates the new field and populates the first two rows with the complete value in column 5 and gives me NaN for the remainder.

These attempts give me syntax erros:

df['new'] = df['column5'].str[0:2]
df.map(lambda df['column5']: [:2])

I am simply at a loss of how to create a new column using the first two digits of an existing column from a table read in via pandas.

If this were SAS I'd have been done hours ago, but I am trying to make a go of Python so your help is appreciated

Answer 1

I guess your column5 column is of int*/float* dtype, so try to convert it to string first:

df['new'] = df['column5'].astype(str).str[:2]

you can explicitly specify types of columns when reading CSV file:

df = pd.read_csv('file_name.csv', ..., dtype={'column5': object})