Is InvalidValueReplacement working?

[wf-r]

Hi,

in the following model, I would assume that InvalidValueReplacement should yield a result different from the one currently produced by PyPmml. Could you have a look at it?

How to reproduce:

import pandas as pd
import pypmml

model = pypmml.Model.fromFile('./out.txt')
pred_data = model.predict(pd.DataFrame([['c'], ['d']], columns=['C']))
assert (pred_data['predicted_T'] == pd.Series([1, 1])).all()  # Value 'd' should be invalid, thus being replaced by 'c'

out.txt

System:

• Python 3.8.6
• PyPmml 0.9.13 (with fix for https://github.com/autodeployai/pypmml/issues/32)

Thank you! Wolfgang

[scorebot]

@wf-r Thanks for your info, it is a bug of PMML4S handling the invalid value. I have fixed it. Please, reinstall the latest version from Github by the following command:

pip install --upgrade git+https://github.com/autodeployai/pypmml.git

Please, let me know if you still have a problem.

[wf-r]

Great, thanks for the quick fix!