scikit-learn：將多輸出決策樹轉換爲CoreML模型

Question

我有一個訓練有素的scikit-learn模型，它使用多輸出決策樹（作爲RandomForestRegressor）。由於內置了多輸出行爲，因此沒有對隨機森林迴歸模型顯式進行自定義配置以啓用多輸出行爲。基本上，只要您將多輸出訓練數據放入模型中，模型將在幕後切換到多輸出模式。

此外，RandomForestRegressor是CoreML轉換腳本提供的支持的轉換器。然而，在轉換過程中，我得到這個錯誤瓦特/堆棧跟蹤：

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. 

簡單的轉換代碼如下：

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") 

我能做些什麼，以使CoreML轉換腳本處理多輸出決策樹？是否可以對現有腳本進行更改，而不用完全重新創建新腳本的輪子？

Answer 1

CoreML是（現在）是一個全新的東西，所以目前沒有任何已知的第三方轉換腳本來源。

coremltools documentation的「模型」部分提供了有關如何使用Python生成CoreML模型的廣泛文檔。話雖如此，您可以使用文檔中提供的模型接口將任何機器學習模型翻譯成CoreML模型。

目前，coremltools不支持多輸出迴歸模型。如果您不希望重新發明輪子，則需要通過引入與當前預測的輸出相對應的新輸入將模型轉換爲單個輸出模型。

無論哪種方式，文檔都在那裏，以便讓你開始。