Difan Deng

Sorry for the late reply, this is truely a bug when working with multi-instances problems, intensifier will not suggest the same configuration when doing intensification, we will figure this out...

Overall it looks good to me @AndreBiedenkapp @mfeurer @KEggensperger any comments?

Hi, sorry for the late response. For your first question: how to let SMAC continue the exact configuration after resuming. Unfortunately I have no concrete idea, this line of code:...

Hi, I created a branch to temporarily solve this problem with the random forests as a surrogate model and hope that it could solve this issue: https://github.com/automl/SMAC3/tree/multi_obj_epm However, it might...

Hi, sorry for the late reply, optimizing the acquisition function values is not parallelized: https://github.com/automl/SMAC3/blob/master/smac/optimizer/ei_optimization.py#L351 DASK is only applied to run your target function parallelly , but not the acquisition...

> When you say "run target function parallelly", does that mean running the target function observation with one config or multiple configs? With multiple configs. However, the current EPM chooser...

> Add ask + tell interface And rename EPMChooser

If everyone is happy with the implementation, I will merge this branch

Generally speaking, EIPS belongs to multi-objective acquisition function and needs to be combined with [multi_objective_model]https://github.com/automl/SMAC3/blob/development/smac/model/multi_objective_model.py We will update the description accordingly

@8W9aG , This error happens if no categorical feature is available. If you are sure that no categorical feature exists in your codebase, you can replace [this line](https://github.com/automl/Auto-PyTorch/blob/master/autoPyTorch/configs/forecasting_init_cfgs.json#L24) with `"feature_encoding:__choice__":...