predict_qd raises ValueError: invalid literal for int() with base 10

[smarie]

This bug happens when one column starts with a nan, and then contains a string. Numpy vectorize is guessing from nan that the mapping operation creates a number, but then it fails when it hits a string.

[smarie]

df = pd.DataFrame({
    "foo": ["1", "2"],
    "bar": [np.nan, "B"]
})
qd_forest = qd_screen(df)
feat_selector = qd_forest.fit_selector_model(df)
only_important_features_df = feat_selector.remove_qd(df)
result = feat_selector.predict_qd(only_important_features_df)