Kamil

@FlorinAndrei @dennisbader I'll do PR.

In LinuxTest python_312_pre -> Run tests (pip pre) an error occurs. In other versions everything works. Looking at List Configuration (PyPI) I saw that the numpy version --- 2.1.0.dev0. The...

What to do with `_var_normal `from `tuning_s_estimator_mean`(test_tuning_smoke) that you mentioned in #9248 (I can check `if norm.continuous == 2:`) or do you have other plans for this? The slowest ones...

The task is not entirely clear, you just need to scale all the numerical data by passing, for example, scale=True to ProfileReport(df, scale=True), that is, create an internal scaling functionality...

@vogt31337 @jthurner Hi! I'm writing to clarify the behavior of `chain assignments` previously used in the codebase. Example: `net[elm].bus[connected_elms] = net[elm].bus[elm1]` As you’ve noted, this line relies on chained indexing,...

I tried to return chain assignments in the `_copy_to_measurement` function by changing the code as follows(chain assignments - commented out): ``` class CreateMeasurements: def __init__(self, net: pandapowerNet, cim: Dict): self.logger...

> > difference is fix2617['element'] is of type int64 in chain assignments, loc float64 > > Did you see the different dtypes when comparing the isolated chained assignment change? Or...

Can do something similar to using chain assignments by using `iloc` instead of `loc:` ``` # get the column index ind_col = df4.columns.get_loc('2') df4.iloc[1:, ind_col] = np.array(df1['1'][:]) ``` I'm not...

> QFT is a probabilistic algorithm, the results will have some variability. I can see that after running each time I'm getting new outputs. Yeah..Let's wait for more feedback. I...

The tests don't pass, I'll try with the original requirements.txt. I tried to return numpy-1.26.4 to requirements.txt, it works.