Cannot export XGBClassifier model: TypeError: unsupported operand type(s) for *: 'int' and 'NoneType'

[git2621]

from sklearn.datasets import load_iris
 
from xgboost.sklearn import XGBClassifier
from xgboost import plot_importance
 
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.metrics import accuracy_score
 
#记载样本数据集
iris = load_iris()
x,y = iris.data,iris.target
 
#数据集分割
x_train,x_test,y_train,y_test = train_test_split(x,y,test_size=0.2,random_state=123457)
 
xgb_clf = XGBClassifier(
    booster = 'gbtree',
    objective = 'multi:softmax',
    num_class = 3,
    gamma = 0.1,
    max_depth = 6,
    reg_lambda = 2,
    subsample = 0.7,
    colsample_bytree = 0.7,
    min_child_weight = 3,
    eta = 0.1,
    seed = 1000,
    nthread = 4,
)
 
#训练模型
xgb_clf.fit(x_train,y_train,eval_metric='auc')
 
import m2cgen as m2c
xgb_clf.base_score = 0
code = m2c.export_to_c(xgb_clf)
with open ('model.c', 'w') as f:
   f.write(code)

Full trace:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[24], line 3
      1 import m2cgen as m2c
      2 xgb_clf.base_score = 0
----> 3 code = m2c.export_to_c(xgb_clf)
      4 with open ('model.c', 'w') as f:
      5    f.write(code)

File ~\AppData\Local\Programs\Python\Python310\lib\site-packages\m2cgen\exporters.py:81, in export_to_c(model, indent, function_name)
     61 """
     62 Generates a C code representation of the given model.
     63 
   (...)
     75 code : string
     76 """
     77 interpreter = interpreters.CInterpreter(
     78     indent=indent,
     79     function_name=function_name
     80 )
---> 81 return _export(model, interpreter)

File ~\AppData\Local\Programs\Python\Python310\lib\site-packages\m2cgen\exporters.py:459, in _export(model, interpreter)
    457 def _export(model, interpreter):
    458     assembler_cls = get_assembler_cls(model)
--> 459     model_ast = assembler_cls(model).assemble()
    460     return interpreter.interpret(model_ast)

File ~\AppData\Local\Programs\Python\Python310\lib\site-packages\m2cgen\assemblers\boosting.py:214, in XGBoostModelAssemblerSelector.assemble(self)
    213 def assemble(self):
--> 214     return self.assembler.assemble()

File ~\AppData\Local\Programs\Python\Python310\lib\site-packages\m2cgen\assemblers\boosting.py:36, in BaseBoostingAssembler.assemble(self)
     34         return self._assemble_bin_class_output(self._all_estimator_params)
     35     else:
---> 36         return self._assemble_multi_class_output(self._all_estimator_params)
     37 else:
     38     result_ast = self._assemble_single_output(self._all_estimator_params, base_score=self._base_score)

File ~\AppData\Local\Programs\Python\Python310\lib\site-packages\m2cgen\assemblers\boosting.py:62, in BaseBoostingAssembler._assemble_multi_class_output(self, estimator_params)
     58 def _assemble_multi_class_output(self, estimator_params):
     59     # Multi-class output is calculated based on discussion in
     60     # https://github.com/dmlc/xgboost/issues/1746#issuecomment-295962863
     61     # and the enhancement to support boosted forests in XGBoost.
---> 62     splits = _split_estimator_params_by_classes(
     63         estimator_params, self._output_size,
     64         self.multiclass_params_seq_len)
     66     base_score = self._base_score
     67     exprs = [
     68         self._assemble_single_output(e, base_score=base_score, split_idx=i)
     69         for i, e in enumerate(splits)
     70     ]

File ~\AppData\Local\Programs\Python\Python310\lib\site-packages\m2cgen\assemblers\boosting.py:347, in _split_estimator_params_by_classes(values, n_classes, params_seq_len)
    342 def _split_estimator_params_by_classes(values, n_classes, params_seq_len):
    343     # Splits are computed based on a comment
    344     # https://github.com/dmlc/xgboost/issues/1746#issuecomment-267400592
    345     # and the enhancement to support boosted forests in XGBoost.
    346     values_len = len(values)
--> 347     block_len = n_classes * params_seq_len
    348     indices = list(range(values_len))
    349     indices_by_class = np.array(
    350         [[indices[i:i + params_seq_len]
    351           for i in range(j, values_len, block_len)]
    352          for j in range(0, block_len, params_seq_len)]
    353         ).reshape(n_classes, -1)

TypeError: unsupported operand type(s) for *: 'int' and 'NoneType'

xgboost version '2.0.3'

[gregstarr]

set num_parallel_tree in XGBClassifier e.g. XGBClassifier(booster="gbtree", tree_method="hist", num_parallel_tree=1)