Missing binary:hinge

[realjanpaulus]

Is there a reason why binary:hinge isn't supported as objective for XGBoost?

onnxmltools/onnxmltools/convert/xgboost/operator_converters/XGBoost.py

Lines 267 to 293 in fe55a8a

	if objective == "binary:logistic":
	ncl = 2
	container.add_node('TreeEnsembleClassifier', operator.input_full_names,
	operator.output_full_names,
	op_domain='ai.onnx.ml',
	name=scope.get_unique_operator_name('TreeEnsembleClassifier'),
	**attr_pairs)
	elif objective in ("multi:softprob", "multi:softmax"):
	ncl = len(js_trees) // params['n_estimators']
	if objective == 'multi:softmax':
	attr_pairs['post_transform'] = 'NONE'
	container.add_node('TreeEnsembleClassifier', operator.input_full_names,
	operator.output_full_names,
	op_domain='ai.onnx.ml',
	name=scope.get_unique_operator_name('TreeEnsembleClassifier'),
	**attr_pairs)
	elif objective == "reg:logistic":
	ncl = len(js_trees) // params['n_estimators']
	if ncl == 1:
	ncl = 2
	container.add_node('TreeEnsembleClassifier', operator.input_full_names,
	operator.output_full_names,
	op_domain='ai.onnx.ml',
	name=scope.get_unique_operator_name('TreeEnsembleClassifier'),
	**attr_pairs)
	else:
	raise RuntimeError("Unexpected objective: {0}".format(objective))