Pierre Bonami

Hi Stefan, I don't think it's really possible to debug your issue and the "Bonmin team" is not so active to be able to do this. From what you show...

What we currently don't support is the [HistGradientBoostingRegressor](https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.HistGradientBoostingRegressor.html#sklearn.ensemble.HistGradientBoostingRegressor), but the monotic constraint is on the training and that is not an issue As far as I can see if you...

Hi, I am not very familiar with Paddle so I wouldn't be able to help much. But in principle there's no objection. Note that it would need to come with...

This is not trivial to do but it can be done by inspecting the modeling object that is created when the neural network is added to the gurobipy model (the...

Doing a very simple example is not so long. What I did doesn't really make sense but I hope it shows how I would do it. Here: https://github.com/pobonomo/gurobi-machinelearning/blob/bound_layers_adversarial/docs/notebooks/ipynb/adversarial_mnist.ipynb it's a...

Yes that would be nice. I am not sure how much I can do in this direction but it would definitely be interesting.

Thinking a bit more about this, there is a documentation effort to make. In most cases those intermediate variables are stored but that part is not documented. This should definitely...

> * Rename the `output_type` parameter to `prediction_method` with the following options, that mostly match the sklearn predictor method names: > > * `predict` > * `predict_proba` > * `decision_function`...

@bzfwunde I removed this predict_class parameter as we had discussed and reworked the Janos example. It has to do some compromises for the example but it's the best I could...