dionis | Dionis predictors | Machine Learning library

dionis is a Python library typically used in Artificial Intelligence, Machine Learning, Deep Learning, Pytorch applications. dionis has no bugs, it has no vulnerabilities, it has a Permissive License and it has low support. However dionis build file is not available. You can download it from GitHub.

This package can be used to construct and blend randomly sampled models for Machine Learning. Only predictors who help to raise final blender AUC score on test set are selected and saved, others skipped. This package can be used for example for Kaggle competitions. At start blender will split X and y to Major train and test sets. Major train set will be used for cross-validation training of all predictors, Major test set will be used to train blender itself. At this point blender will generate Cross Validation (CV) folds and randomly selects one of predefined models, then it will randomly sample parameters of model distribution and start fitting model to targets. After CV procedure is finished, it will save models in save_path directory.