whether they survived, etc. Let’s try to perform One Hot Encoding on the embark town column.

Let’s first read the dataset, fill in the missing values in the embark town column with the most frequent value of the column, and then, print the number of passengers from each embark town.

import seaborn
from sklearn.preprocessing import OneHotEncoder

df = seaborn.load_dataset("titanic")

print("Value counts before filling missing values: \n", df.embark_town.value_counts())

df["embark_town"].fillna(value="Southampton", inplace=True)

print("Value counts after filling missing values: \n", df.embark_town.value_counts())

Here, we are filling in the missing values in the embark town column with the value “Southampton”, which is the most frequent value of the column. The value_counts() function before and after filling in the missing values will look like the following:

Value counts before filling in missing values: 
Southampton    644
Cherbourg      168
Queenstown      77
Name: embark_town, dtype: int64
Value counts after filling missing values: 
Southampton    646
Cherbourg      168
Queenstown      77
Name: embark_town, dtype: int64

Now, let’s do the One-Hot Encoding.

import seaborn
from sklearn.preprocessing import OneHotEncoder

df = seaborn.load_dataset("titanic")

print("Value counts before filling missing values: \n", df.embark_town.value_counts())

df["embark_town"].fillna(value="Southampton", inplace=True)

print("Value counts after filling missing values: \n", df.embark_town.value_counts())

one_hot_encoder = OneHotEncoder()
transformed_df = one_hot_encoder.fit_transform(df[["embark_town"]])
print(type(transformed_df))
print(transformed_df.toarray())

Here, we are using the OneHotEncode class from the sklearn.preprocessing module to perform the One Hot Encoding. The …