Add to Favourites
To login click here
Reverb is an efficient and easy-to-use data storage and transport system designed for machine learning research. It is primarily used as an experience replay system for distributed reinforcement learning algorithms, but also supports multiple data structure representations such as FIFO, LIFO, and priority queues. Reverb is not hardened for production use and only supports Linux based OSes. It can be installed with pip, and TensorFlow can be installed separately or as part of the install. Experience replay has become an important tool for training off-policy reinforcement learning policies.
