Nicolaj Søndergaard Mühlbach

ImportError: /lib64/libm.so.6: version `GLIBC_2.29' not found

6

Hi there! First, I love this library! However, I'm recently facing issues loading the library. I'm running on a Linux server with ``` Python 3.8.8 [GCC 7.3.0] :: Anaconda, Inc....

Add option to specify "catch" in OptunaSearchCV

1

### Expected behavior I'm hitting this error: `FloatingPointError: underflow encountered in _ndtri_exp_single (vectorized)`, when I'm fitting an `OptunaSearchCV` instance with a XGBoostRegressor, where regularization parameters are sampled too low (I...

Sampler getting stuck non-optimally in search space?

2

### Expected behavior It seems that the `TPESampler` can get stuck and repeatedly sample parameters that result in `NaN`/`None` without ability to move away from these parameters. I cannot reproduce...

What is the point of "nonconformity_quantiles" when "quantiles" can be specified independently at inference?

4

Hi maintainers! Love your work! I'm curious if you could explain what the argument `nonconformity_quantiles` does and if it's important to get could predictions of the quantiles? How does it...

Arg `always_select` does not seem to strictly enforce non-zero coefficients

4

Thanks for a great library! When using the arg `always_select`, I expect that the model will always estimate at least `len(always_select)` non-zero coefficients. However, this is not the case (unless...