salesforce/TransmogrifAI

TransmogrifAI is a machine learning framework built on Apache Spark that automates feature engineering, model selection, and evaluation. It provides a comprehensive suite of transformers, estimators, and evaluators for classification, regression, and forecasting tasks, alongside utilities for data ingestion and local scoring. The system supports diverse data types and includes tools for model interpretability and streaming analysis.

This period marks the release of version 0.7.0, which introduces support for Apache Spark 2.4.5 and rebrands the project as TransmogrifAI. The release adds new machine learning models, feature types, and data readers, while also expanding the CLI and test coverage for the new functionality.

47 changes

This period focused on expanding test coverage across core, features, and utility modules, alongside the introduction of local scoring capabilities and streaming histogram utilities. The work also included adding Jupyter notebook examples and implementing custom serialization annotations to support in-memory model evaluation and legacy model format testing.

17 changes

salesforce/TransmogrifAI was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point the surveyor at a repository you know and see whether you agree with it.