Sometimes in datasets, we encounter columns that contain numbers of no specific order of preference. The data in the column usually denotes a category or value of the category and also when the data in the column is label encoded. This confuses the machine learning model, to avoid this the data in the column should be One Hot encoded.
One Hot Encoding –
It refers to splitting the column which contains numerical categorical data to many columns depending on the number of categories present in that column. Each column contains “0” or “1” corresponding to which column it has been placed.
For example :
Consider the data where fruits and their corresponding categorical value and prices are given.
|Fruit||Categorical value of fruit||Price|
The output after one hot encoding the data is given as follows,
Below is the Implementation in Python –
The following example is the data of zones and credit scores of customers, the zone is a categorical value which needs to be one hot encoded.
To one hot encode the zone column –
The output contains 5 columns, one column for the price, and the remaining 4 columns representing the 4 zones.
One hot encoder only takes numerical categorical values, hence any value of string type should be label encoded before one-hot encoded.
The below example has the data of geography and gender of the customers which has to be label encoded first.
Label encoding the data –
One Hot Encoding Gender and Geography Columns –
The output contains 5 columns, 2 columns representing the gender, male and female, and the remaining 3 columns representing the countries France, Germany, and Spain.
- The one hot encoder does not accept 1-dimensional array or a pandas series, the input should always be 2 Dimensional.
- The data passed to the encoder should not contain strings.