Create hyperparameter for block update in BoostAODE #25

New Issue

rmontanana · 2024-04-08T20:50:19Z

rmontanana commented

2024-04-08 20:50:19 +00:00

Create a hyperparameter block_update to perform weights updating and significance computing per block and not per model as it is done by now.
This way if bisection is enabled when inserting for instance 8 models the weights and the significance of these 8 models shall be computed once the whole 8 are inserted to the ensemble, and the predict to compute the α_t, ε, etc. is done taking a micro ensemble with the models belonging to that block.

Create a hyperparameter *block_update* to perform weights updating and significance computing per block and not per model as it is done by now. This way if bisection is enabled when inserting for instance 8 models the weights and the significance of these 8 models shall be computed once the whole 8 are inserted to the ensemble, and the predict to compute the α<sub>t</sub>, ε, etc. is done taking a micro ensemble with the models belonging to that block.

rmontanana added this to the v1.0.5 milestone 2024-04-08 20:50:19 +00:00

rmontanana added the

💡 enhancement

label 2024-04-08 20:50:19 +00:00

rmontanana self-assigned this 2024-04-08 20:50:20 +00:00

rmontanana commented

2024-05-02 09:06:08 +00:00

Already implemented !26

rmontanana closed this issue

2024-05-02 09:06:09 +00:00

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: rmontanana/BayesNet#25