Articles on: Training models

Model summary

Model Summary displays the accuracy of your model. Accuracy in machine learning refers to the percentage of correct predictions made by a model on a dataset, and it is a classification metrics. Improving accuracy involves finding ways to make the model better at predicting correct outcomes. This can be achieved through several methods, including:

Collecting more training data.
Using more advanced algorithms or models.
Feature Engineering (i.e. creating new features or modifying existing features to improve model performance).
Data preprocessing and cleaning.

You can have a quick grasp about your model's performance through the Model summary panel:

A Sample Model Summary Graph and Suggestions

You will also see how much your model performed against a simpler baseline method. For example, if a baseline method has an accuracy of 80% and the new method has an accuracy of 90%, then the new method is said to be "10% better" or "1.125x better" than the baseline method.

Keep in mind that accuracy is only one way to evaluate the model's performance, and you need to also take other model performance metrics into account. You can learn more about model evaluation metrics through this NextBrain article: https://help.nextbrain.ai/en/article/advanced-statistics-1skca2c/?bust=1675184664824

Updated on: 02/02/2023

Was this article helpful?

Share your feedback

Cancel

Thank you!