CODEWITHSUNDEEP
pythonpandas
sayan nandi
sayan nandi

Reputation: 83

Filtering a column for only Numerical values in Pandas

I have a dataframe where a column is named as USER_ID. Ideally USER_ID should be of numerical No but the data that is coming from source is having typically some bad records which i want to discard in my final dataframe. For example the values in the column are like below

DF

USER_ID
23456
1236
NO_NULL
FBA56X%^

and the final dataframe should be

DF1

USER_ID
23456
1236

The code i am using to clean it is as below

DF1 = DF[np.isfinite(pd.to_numeric(DF.USER_ID,errors='coerce))]

But it seems this code is not acting properly. Any suggestion will be appreciated.

Upvotes: 0

Views: 7094

Answers (1)

Nk03
Nk03

Reputation: 14949

You can use isnumeric() to filter the numeric values:

df1 = df.loc[df.USER_ID.str.isnumeric()]

Upvotes: 2

Related Questions