Python (Lab) Error

[KNIMEuser23421]

Python code:

This Beta.ppf line from my VS Code returns following error messages:

• Miniconda3\lib\site-packages\scipy\stats_continuous_distns.py:692: RuntimeWarning: Error in function boost::math::tgamma(d,d): Series evaluation exceeded %1% iterations, giving up now.
  return _boost._beta_ppf(q, a, b)
  -Miniconda3\lib\site-packages\scipy\stats_continuous_distns.py:692: RuntimeWarning: divide by zero encountered in _beta_ppf
  return _boost._beta_ppf(q, a, b)

But these are warnings, and the code still runs.

however the python(labs) does not execute:

In R I get warnings too:

What I am trying to do:
I am trying to get Beta variable given quantile and shape parameter alpha and beta. But the are some abnormal values in the data which will return wacky alpha and beta combination. So both R and VS Code will give me warnings but it runs. However, the same code in the Python(lab) node will not execute and will return error.

I need help on:

1. Why python(lab) will not run. is there a fix on this ?
2. This is more for statisticians, How would I determine what kind of alpha and beta combination will give me warning messages? If I know the extreme values for alpha and beta, think I can put some restrictions on it so that it will run within both R and VS Code without warning → therefore will execute within python(lab) .

Many Thanks for helps and tips!

[steffen_KNIME]

Hi KNIMEuser23421,

did you test the code also in the Python Script node (without labs)? Did you test it in a normal Python environment? I would like to exclude the possibility that it is a problem of Python packages and confirm that it is a bug of our Python Script (Labs) node.

Best regards
Steffen

[mlauber71]

@KNIMEuser23421 maybe you could provide a code example to test.