Initializing the Classification Environment

Learn to initialize the classification environment for natural language processing task.

We'll cover the following

Classification environment setup

Now that we have a dataset with 5 numeric features and a class label, we can train a classification model on it. By doing that, we’ll be able to create a text classification pipeline using the topic and classification models. At this point, we’re dealing with a multiclass classification task. We’ll discuss that more briefly.

First of all, we initialize the PyCaret classification environment that prepares the dataset for model training. During EDA, we discovered that the dataset was imbalanced, so we set the fix_imbalance parameter to True.

After running setup(), we get a table of useful information about its settings and parameters, explained in detail below.

Note: After running the code, we can see the table, given below, by scrolling up in the terminal.

Get hands-on with 1200+ tech skills courses.