insolver 0.4.19

Insolver is low-code machine learning library, initially created for the insurance industry.

Insolver

Insolver is a low-code machine learning library, originally created for the insurance industry, but can be used in any other. You can find a more detailed overview here.

Installation:

pip install insolver

Insolver is already installed in the easy access cloud via the GitHub login. Try https://mset.space with a familiar notebook-style environment.

Examples:

• Binary Classification Example - Rain in Australia Prediction This tutorial demonstrates how to create classification models for the weatherAUS dataset: getting and preprocessing data, transformations, creating models, plotting SHAP values and comparing models.

• Data Preprocessing Example I - New York City Airbnb This tutorial demonstrates how to use the feature_engineering module and all the main features of each class. For this, the AB_NYC_2019 dataset is used.

• Data Preprocessing Example II - New York City Airbnb This tutorial also demonstrates how to use the feature_engineering module, but it covers the automated data preprossesing class and all of its features. For this, the AB_NYC_2019 dataset is used.

• Gradient Boosting Example - Lending Club This tutorial demonstrates how to create classification models for the Lending Club dataset using the Gradient Boosting libraries and the InsolverGBMWrapper class.

• Inference Example This tutorial demonstrates how to load InsolverTransform transforms from a file and initialize them using the init_transforms function.

• InsolverDataFrame and InsolverTransform Example This tutorial demonstrates main features of the InsolverDataFrame class and the InsolverTransform class.

• Regression Example - FreeMLP This tutorial demonstrates how to create regression models for the freMPL-R dataset: getting and preprocessing data, transformations, creating models, plotting SHAP values and comparing models.

• Regression Example - US Accidents This tutorial demonstrates how to create regression models for the US Traffic Accident dataset: getting and preprocessing data, transformations, creating models, plotting SHAP values and comparing models.

• Report Example This tutorial demonstrates how to create a HTML report with different models using the Report class.

Documentation:

Available here

Supported libraries:

Libs:

• sklearn
• H2O

Boosting models:

• XGBoost
• LightGBM
• CatBoost

Model interpretation:

• shap plots

Serving (REST-API):

• flask
• fastapi

Run tests:

python -m pytest

tests with coverage:

python -m pytest --cov=insolver; coverage html; xdg-open htmlcov/index.html

Contributing to Insolver:

Please, feel free to open an issue or/and suggest PR, if you find any bugs or any enhancements.

Demo

Example of creating models using the Insolver

Example of a model production service

Example of an elyra pipeline built with the Insolver inside

Contacts

[e-mail] [phone]