BayesianOptimization

BayesianOptimization copied to clipboard

KeyError: 'Data point X is not unique'

11

Hi and congrats for this sweet piece of software, really love it :) I've implemented an async solution very similar to the one in the example notebook, with 13 virtual...

PoCk3T

log files are not valid json format

3

It is a small issue, but annoying non the less. When I try to commit a json log file, it gets flagged by the [json5 pre-commit hook](https://gitlab.com/bmares/check-json5 ). ``` check...

rruiter87

Same observation being generatad

10

Hi, I tried to run the code below to optimize a XGBoost classifier, but get stuck with same observation being tested all time. I expected some new observation being generated......

brunodoamaral

Better sampling of the random exploration step using Latin hypercube

2

Choosing the initial exploration points as independent random variables is not optimal for achieving high coverage of the parameter space. One better solution is using a Latin hypercube. This could...

MariusCautun

Formatting issues in Windows

5

Hello, when executed in Windows, the optimization have some weird formatting errors. Like this: | iter | target | bits | size | ------------------------------------------------- |  1  | ...

LucaCerina

Handling of parameters with integer values

6

Treating integer-valued parameters as floats can be sub-optimal for the maximization procedure. This is discussed in https://arxiv.org/abs/1706.03673, where the authors have shown that accounting for the integer nature of some...

MariusCautun

The target value in the logs is wrong

4

The 18th trial shows a target of -1.826e+0. But at the bottom you can see that the actual target value is -182571.40432067157. It's wrong by several orders of magnitude. 

outdoteth

Support different data types for optimization parameters

15

It would be nice to support different data types---e.g. `int`, `float`, `bool`, and maybe a categorical string---for the parameters over which we optimize. I am not sure what the syntax...

Engineero

mismatch between the next suggestion point (optimizer.suggest(utility)) and the argmax(EI) value

1

Hello. Thanks for your bayesian optimization codes, it's really helpful. I have a question in a suggestion mechanism. When I run the code, there was a next suggestion point mismatch,...

shoh4486

Using enums instead of string is easier for command completion.  Which avoids the error when typing 'pi' instead of the correct 'poi'. This implementation accepts both enum and strings...

ecomodeller