Léo Géré

I like the possibility to be able to directly run the python file to render the scene without the need to use the command line. Most IDEs provide a simple...

I don't know if this helps in any way, but I usually add the following line at the very end of my function to build the model, just before returning...

Having a flag to disable keeping checkpoints for finished trials would be very helpful. I don't need to keep the trained model, I just need the best HP combination that...

These methods would be very extremely useful!

This would be very useful

Any progress on this algorithm?

I would be very interested in this being fixed too! Being able to compare models using slightly different sets of hyperparameters is very important in my experiments.

This method almost always returns the default value of the nested HP if the search space of the outer HP is large. It only works when the outer HP takes...

Just an example in which this functionality would be useful: Let's say I need to construct a fully-connected network, and I have a maximum budget of 1M parameters (for whatever...

This feature would be very interesting, thank you for your work. As I understand the discussion, this could eventually open the possibility (in a future PR) to manually draw bounding...