Converter for pyod.models.feature_bagging.FeatureBagging

[hanzigs]

Is it possible to get converter and shape calculator for pyod feature bagging? https://pyod.readthedocs.io/en/latest/_modules/pyod/models/feature_bagging.html

import pandas as pd
from pyod.models.feature_bagging import FeatureBagging
from sklearn.preprocessing import MinMaxScaler
from onnxmltools.convert.common.data_types import FloatTensorType
from skl2onnx import to_onnx

data = {'First':  [500,500,400,100,200,300,100,500,500,400,100,200,300,100,200,300,100,500,500,400,100,200],
         'Second': ['a','b','a','b','a','b','c','a','b','a','b','a','b','c','a','b','c','a','b','a','b','c']}
df = pd.DataFrame(data,columns=['First','Second'])
dumdf = pd.get_dummies(df) 
scaler = MinMaxScaler()
scaler.partial_fit(dumdf)
sc_data = scaler.transform(dumdf)
# pip install combo
fb_clf = FeatureBagging(n_estimators=10, contamination=0.1).fit(sc_data)
feature_names = dumdf.columns

initial_type = [('float_input', FloatTensorType([None, len(feature_names)]))]
FBOnnx_model = to_onnx(fb_clf, initial_types=initial_type)

I understand there are lot of algorithms where all of them cannot be supported, Just would like to know the possibility, Thank you very much

[xadupre]

Sorry for the later answer. FeaureBagging is outside ot scikit-learn and that's out of the library scope. It could be possible to add it to the documentation as an example. It could help if you would have a function written in numpy equivalent to FeatureBagging.decision_function.