Evaluate your optimizer
In order to evaluate your optimizer, you can run:
python evaluate.py --problem=$problem_name --optimizer=L2L --path=$path_to_the_saved_model
where $problem_name
is the name of your problem; --optimizer
gives the name of the optimizer and --path
is the path to your trained model. For instance, in order to evalute a 2D Rastrigin function, you can set --problem=square_cos_2d
and --path=../trained_models/square_cos_2/
.