Is your feature request related to a problem? Please describe.
Tutorial 5 is currently very similar to Tutorial 1 (essentially the same workflow, with ANN instead of RBF) and also overlaps significantly with Tutorial 2, which already provides a comparison of the available methods in EZyRB. As a result, it does not add substantial new information for the user.
A more valuable approach would be to highlight and explain the tunable parameters within the ANN module, giving users a clearer understanding of how to customize and control the model.
Describe the solution you'd like
Refactor Tutorial 5 to focus on:
the configuration and tuning of ANN parameters,
practical guidance on how these parameters affect performance,
use cases where ANN behaves differently from other approximators.
Describe alternatives you've considered
Additional context
Even when addressing a specific request for a new tutorial, the goal should not simply be to close the issue, but to meaningfully expand the knowledge provided to users.
If the only difference from existing tutorials is a minor change (e.g., switching one approximator), the learning outcome remains essentially the same. While users may notice that approximators can be swapped, this is already clearly conveyed in Tutorial 2.
Therefore, the focus should be on delivering genuinely new insights rather than minimal variations of existing content.
Is your feature request related to a problem? Please describe.
Tutorial 5 is currently very similar to Tutorial 1 (essentially the same workflow, with ANN instead of RBF) and also overlaps significantly with Tutorial 2, which already provides a comparison of the available methods in EZyRB. As a result, it does not add substantial new information for the user.
A more valuable approach would be to highlight and explain the tunable parameters within the ANN module, giving users a clearer understanding of how to customize and control the model.
Describe the solution you'd like
Refactor Tutorial 5 to focus on:
the configuration and tuning of ANN parameters,
practical guidance on how these parameters affect performance,
use cases where ANN behaves differently from other approximators.
Describe alternatives you've considered
Additional context
Even when addressing a specific request for a new tutorial, the goal should not simply be to close the issue, but to meaningfully expand the knowledge provided to users.
If the only difference from existing tutorials is a minor change (e.g., switching one approximator), the learning outcome remains essentially the same. While users may notice that approximators can be swapped, this is already clearly conveyed in Tutorial 2.
Therefore, the focus should be on delivering genuinely new insights rather than minimal variations of existing content.