Dockerfile.tar.gz

Numerical Variables

Default Job

In this course, you will learn how to apply basic and more advanced feature engineering to tabular data with python, during this course,  we will see a range of different techniques and methods to handle many common cases within the data set, so in result, you will create great features so that your machine learning models can predict good results.

Feature Engineering with Python

## Categorical variable — cardinality

The values of a categorical variable are selected from a group of categories, also called labels. The number of unique labels is called __cardinality__.

Let us consider the example of having an interview for a job position, and the variable would be the result of this interview. The result can only be accepted or refused, which means we have two labels, so the cardinality of this variable is two (2).

## Cardinality impact
However, cardinality or multiple labels in a categorical variable can impact our model. Here are a few reasons why:
- We need to encode those labels into numbers. Categories must be represented numerically, and encoding techniques affect feature space and variable interactions.
- There can be an unbalanced distribution between the train and test sets. For highly cardinal variables, some labels may appear only in the train set, leading to over-fitting.
Some labels may appear only in the test set, which means the model does not know these labels and will not interpret them.
- Variables with too many labels tend to dominate over those with fewer labels, specifically for tree-based algorithms. If a variable has many labels, it may cause noise in the dataset, so reducing cardinality can improve the model's performance.

# Categorical variable — cardinality

The values of a categorical variable are selected from a group of categories, also called labels. The number of unique labels is called __cardinality__.

Let us consider the example of having an interview for a job position, and the variable would be the result of this interview. The result can only be accepted or refused, which means we have two labels, so the cardinality of this variable is two (2).

# Cardinality impact
However, cardinality or multiple labels in a categorical variable can impact our model. Here are a few reasons why:
- We need to encode those labels into numbers. Categories must be represented numerically, and encoding techniques affect feature space and variable interactions.
- There can be an unbalanced distribution between the train and test sets. For highly cardinal variables, some labels may appear only in the train set, leading to over-fitting.
Some labels may appear only in the test set, which means the model does not know these labels and will not interpret them.
- Variables with too many labels tend to dominate over those with fewer labels, specifically for tree-based algorithms. If a variable has many labels, it may cause noise in the dataset, so reducing cardinality can improve the model's performance.

In this lesson, we will see a very important concept when it comes to categorical variables called "cardinality", what it is and how it can affect the performance of our model.


Categorical variable — cardinality

In this lesson, we will see a very important concept when it comes to categorical variables called "cardinality", what it is and how it can affect the performance of our model.

Introduction

Variable Types

Common Concerns in Datasets

Handling & Imputing Missing Values

Encoding Categorical Variables

Transforming Variables

Variable Discretization

Handling Outliers

Feature Scaling

Engineering Geospatial Data

Handling Date-Time and Mixed Variables

Resampling Imbalanced Data

Advanced Feature Engineering Techniques

Conclusion

Categorical variable — cardinality

Categorical variable — cardinality