A modular RL library to fine-tune language models to human preferences. It provides easily customizable building blocks for training language models, including implementations of on-policy algorithms, reward functions, metrics, datasets, and actor-critic policies.  

A toolkit to develop RL agents for common NLP tasks such as sequence tagging, multi-label classification and multiple-choice question answering.

FluidML is a lightweight framework for developing machine learning pipelines. It helps to develop ML pipelines fluently with no boilerplate code. 

A lightweight tool to manage and track your large-scale machine-learning experiments. DashifyML contains two components: a multiprocessing capable logging component and a web-based visualization component.