Trying to export MultiOutputRegressor(LGBMRegressor()) to onnx

[jaegerjj]

Hi,

I tried to export this model to onnx:

from sklearn.multioutput import MultiOutputRegressor from lightgbm import LGBMRegressor model = MultiOutputRegressor(LGBMRegressor()) model.fit(X_train, y_train)

import onnxmltools import onnxmltools.convert.common.data_types from skl2onnx import convert_sklearn from skl2onnx.common.data_types import FloatTensorType initial_type = [('float_input', FloatTensorType([None, 1997]))] onx = onnxmltools.convert.convert_lightgbm(model, initial_types=initial_type) with open("test.onnx", mode = 'wb') as f: f.write(onx.SerializeToString())

https://prnt.sc/tbSL1q9JFmzs

Output: ValueError: No proper operator name found for '<class 'sklearn.multioutput.MultiOutputRegressor'>'

Is it possible to use onnxmltools to export this model to onnx? Thanks for any help.