Portfolio Allocation with TensorTrade: (Part 2/2)
Published in
5 min readApr 12, 2021
In the previous article, I went over how to create a trading environment using the generic components provided in the TensorTrade library. This article will focus on how to tune, train, and evaluate an agent over the built environment while also showing how to make custom models and action distributions in Ray. Without further delay, let’s start with building the model and action distribution!
Ray Components
When working with RL environments, it's convenient to customize different components of the training…
