Reinforcement Learning agent to train real life lifters
more info will be posted when it is ready to use
todo eventually---switch over to single timeseries channel for everything dont have day + workouts in their own timeseries channel todo eventually---make rl agent training distributed todo eventually---somehow put that it took a NEXT_DAY action in the timeseries --i dont think the NN is figuring out that when u add a day something happens? maybe it is but this would make it more explicit kind of like sentence modeling there is a period