Mdatagen

I'm excited to announce a new release of the **mdatagen** package! This update includes:

- **Bug Fixes**: Resolved issues in the EvaluationImputation class. You can now import and use this class correctly. Examples have been added to the documentation for your convenience.
- **Optimized Performance**: Enhanced the multivariate MAR and MNAR classes for better efficiency.
- **Parallelization Added**: Implemented a parallelization strategy for multivariate MAR and MNAR, enabling the generation of missing data in large datasets more efficiently.

Check out the updated docs and give the new features a try! 🚀

Fix Bugs in MNAR mechanism in both univariate and multivariate scenarios

I am excited to announce the new release of my Python library, featuring bug fixes and improvements to the **MAR.correlated** function.

**Warning**: The Bayesian approach from method MBIR in MNAR multivariate.

We want to inform users that at the time of the first release, the Bayesian approach of the MBIR method is not yet implemented. We are actively working to include this functionality in future versions. We thank you for your understanding and patience.

For more information and updates, see the documentation or release notes.

**Warning**: The Bayesian approach from the MBIR method in MNAR multivariate has not been implemented yet.

We have fixed bugs in the _median_ function for MNAR multivariate and mMCAR.

Now, users have the option to choose the generation of synthetic data within the target class for the multivariate scenario. If set to True, when instantiating mMAR, mMNAR, and mMCAR, you must pass the parameter missTarget=True. The default value is missTarget=False.

We also reviewed the code, and now the mdatagen library is functional on Google Colab.

For more information and updates, please refer to the documentation or release notes.