Aerosolve vs baikal: What are the differences?

What is Aerosolve? A machine learning package built for humans (created by Airbnb). This library is meant to be used with sparse, interpretable features such as those that commonly occur in search (search keywords, filters) or pricing (number of rooms, location, price). It is not as interpretable with problems with very dense non-human interpretable features such as raw pixels or audio samples.

What is baikal? A graph-based functional API for building complex scikit-learn pipelines. It is a graph-based, functional API for building complex machine learning pipelines of objects that implement the scikit-learn API. It is mostly inspired on the excellent Keras API for Deep Learning, and borrows a few concepts from the TensorFlow framework and the (perhaps lesser known) graphkit package. It aims to provide an API that allows to build complex, non-linear machine learning pipelines.

Aerosolve and baikal belong to "Machine Learning Tools" category of the tech stack.

Some of the features offered by Aerosolve are:

• A thrift based feature representation that enables pairwise ranking loss and single context multiple item representation.
• A feature transform language gives the user a lot of control over the features
• Human friendly debuggable models

On the other hand, baikal provides the following key features:

• Build non-linear pipelines effortlessly
• Handle multiple inputs and outputs
• Add steps that operate on targets as part of the pipeline

Aerosolve and baikal are both open source tools. It seems that Aerosolve with 4.62K GitHub stars and 583 forks on GitHub has more adoption than baikal with 553 GitHub stars and 23 GitHub forks.