develop

[jzsmoreno]

Pull Request Description

Overview

This pull request introduces several key updates and improvements to the likelihood Python package, including the addition of a new GetInsights class, updates to the autoencoders.py module, enhancements to the AutoClassifier class, and improvements to the Deep Models Jupyter notebook. These changes aim to improve functionality, usability, and performance.

Changes Made

• Add GetInsights class:
  Introduced a new class, GetInsights, to provide deeper insights and analysis related to the model's predictions and performance. This class offers various methods to extract useful statistics and visualizations from the model outputs.

• Update autoencoders.py:
  Refactored and enhanced the existing autoencoders.py module to improve the handling of certain model parameters, optimizing the autoencoder's training process and performance.

• Add _statistics method:
  Added a new private method, _statistics, to the relevant class(es) for better handling of statistical data. This method provides more comprehensive data processing capabilities.

• Add new arguments to AutoClassifier:
  Extended the AutoClassifier class with additional arguments, allowing for more customization and flexibility during model instantiation and training.

• [FIX] GetInsights class:
  Fixed various issues in the GetInsights class that were affecting its functionality. This includes bug fixes related to how insights were computed and presented.

• Update Deep_Models.ipynb:
  Made updates to the Deep_Models.ipynb Jupyter notebook, reflecting the latest changes in the codebase and ensuring that the notebook demonstrates the new features and improvements.

Motivation

The changes in this pull request address several user requests and improve the overall structure of the package, providing better tools for understanding model performance and making the codebase more maintainable and extensible.

Impact

• New features like the GetInsights class and the improvements to AutoClassifier provide additional flexibility and insights.
• Bug fixes and refactorings enhance the stability and usability of the package.
• The updates to the notebook ensure that the latest features are showcased and easy to use.