Training agent using reinforcement learning

Please, how can I train agent using reinfircement learning that manage energy in smart grid for one year because make the episode for one year is very large and make the training impossible.

Answer 1

[Emmanouil Tzorakoleftherakis]

Hello,

When you train using historical data, it is often a good idea to break down your dataset in smaller pieces. Then, instead of training for a year's worth of data, you can make the episode run, e.g., for a month's worth of data. To make sure you use your entire dataset, switch between your smaller datasets using the reset function.

Hope that helps