iso 0.1.0

Modular data transformations for machine learning workflows.

iso

iso is a python package aimed at simplifying the development and deployment cycle associated with building machine-learning models. It enables much more mental clarity during the process of designing a learning algorithm by allowing developers to easily compartmentalize the problem they’re working on.

Features:

• Ability to define modular data transformations that become a part of your model. This helps developers stay organized during the model building process, and makes transitioning models across developers painless.

• Tools for transforming both predictors AND responses in a single step.

• Tools for easily doing data augmentation.

• Ability to easily save and deploy models.

• Allows for more intuitive interaction with model building process.

Comprehensive documentation is currently lacking, but will be included in the next release.

Documentation

For installation and usage instructions please see the documentation.

Questions/Feedback

File an issue in the GitHub issue tracker.