Modeller vs. Keepsake

In the current era of machine learning, software designed for model development leverages over three decades of expertise in credit risk modeling. Modeller stands out as a versatile, clear, and interactive tool that empowers organizations to maximize the potential of their analytical teams. It provides a wide range of methodologies, facilitates the swift creation of robust models, ensures comprehensive explanations, and nurtures the growth of junior team members. Users can select from diverse modeling approaches, including machine learning, to attain the highest levels of predictive precision, particularly when dealing with intricate relationships and multicollinearity. With a simple click, one can generate standard industry binary and continuous target models. The software supports decision tree modeling through CHAID trees and CART methods, along with options like logistic regression, elastic net models, survival analysis (Cox PH), random forest, XGBoost, and stochastic gradient descent. Furthermore, it offers export capabilities to SAS, SQL, and PMML, enabling seamless integration with other scoring and decision-making applications. This flexibility ensures that organizations can easily adapt the models to fit their specific operational contexts and requirements.

Keepsake is an open-source Python library tailored for overseeing version control within machine learning experiments and models. It empowers users to effortlessly track vital elements such as code, hyperparameters, training datasets, model weights, performance metrics, and Python dependencies, thereby facilitating thorough documentation and reproducibility throughout the machine learning lifecycle. With minimal modifications to existing code, Keepsake seamlessly integrates into current workflows, allowing practitioners to continue their standard training processes while it takes care of archiving code and model weights to cloud storage options like Amazon S3 or Google Cloud Storage. This feature simplifies the retrieval of code and weights from earlier checkpoints, proving to be advantageous for model re-training or deployment. Additionally, Keepsake supports a diverse array of machine learning frameworks including TensorFlow, PyTorch, scikit-learn, and XGBoost, which aids in the efficient management of files and dictionaries. Beyond these functionalities, it offers tools for comparing experiments, enabling users to evaluate differences in parameters, metrics, and dependencies across various trials, which significantly enhances the analysis and optimization of their machine learning endeavors. Ultimately, Keepsake not only streamlines the experimentation process but also positions practitioners to effectively manage and adapt their machine learning workflows in an ever-evolving landscape. By fostering better organization and accessibility, Keepsake enhances the overall productivity and effectiveness of machine learning projects.