scikit-learn：将多输出决策树转换为 CoreML 模型

Question

我有一个训练有素的 scikit-learn 模型，它使用多输出决策树（作为 RandomForestRegressor）。没有对随机森林回归模型明确进行自定义配置以启用多输出行为，因为多输出行为是内置的。基本上，只要你给模型拟合多输出训练数据，模型就会在后台切换到多输出模式。

此外，RandomForestRegressor 是 CoreML 转换脚本提供的受支持转换器。但是，在转换过程中，我收到带有堆栈跟踪的错误：

ValueError: Expected only 1 output in the scikit-learn tree.

Traceback (most recent call last):
  File "/Users/user0/Desktop/model_convert.py", line 7, in <module>
    coreml_model = sklearn_to_ml.convert(model)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_converter.py", line 146, in convert
    sk_obj, input_features, output_feature_names, class_labels = None)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_converter_internal.py", line 297, in _convert_sklearn_model
    last_spec = last_sk_m.convert(last_sk_obj, current_input_features, output_features)._spec
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_random_forest_regressor.py", line 53, in convert
    return _MLModel(_convert_tree_ensemble(model, feature_names, target))
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 195, in convert_tree_ensemble
    scaling = scaling, mode = mode, n_classes = n_classes, tree_index = tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 68, in _recurse
    _recurse(coreml_tree, scikit_tree, tree_id, left_child_id, scaling, mode, n_classes, tree_index)
  File "/Library/Python/2.7/site-packages/coremltools/converters/sklearn/_tree_ensemble.py", line 75, in _recurse
    raise ValueError('Expected only 1 output in the scikit-learn tree.')
ValueError: Expected only 1 output in the scikit-learn tree.

简单的转换代码如下：

from coremltools.converters import sklearn as sklearn_to_ml
from sklearn.externals import joblib

model = joblib.load("ms5000.pkl")

print("Converting model")
coreml_model = sklearn_to_ml.convert(model)

print("Saving CoreML model")
coreml_model.save("ms5000.mlmodel")

如何使 CoreML 转换脚本能够处理多输出决策树？我是否可以对现有脚本进行更改，而无需使用新脚本完全重新发明轮子？

Answer 1

CoreML（现在）是一个全新的东西，因此目前没有任何已知的第三方转换脚本来源。

coremltools documentation 的 "Models" 部分提供了有关如何使用 Python 生成 CoreML 模型的大量文档。也就是说，您可以使用文档中提供的模型接口将任何机器学习模型转换为 CoreML 模型。

目前，coremltools 不支持多输出回归模型。如果您不想重新发明轮子，则需要通过引入与当前预测的输出相对应的新输入，将模型转换为单一输出模型。

无论如何，这里有文档，可以帮助您入门。

scikit-learn：将多输出决策树转换为 CoreML 模型

scikit-learn: Convert multi-output decision tree to CoreML model

python

machine-learning

scikit-learn

coreml

coremltools