当使用sklearn2pmml将模型转换为PMML时，sklearn.preprocessing._data.MinMaxScaler不是受支持的转换器

Question

from sklearn.preprocessing import StandardScaler,MinMaxScaler
from sklearn2pmml.pipeline import PMMLPipeline
from sklearn.impute import SimpleImputer
from sklearn_pandas import DataFrameMapper

mapper = DataFrameMapper([
        (model_col,[SimpleImputer(strategy = "most_frequent"),MinMaxScaler()])
    ])

clf = GradientBoostingClassifier(n_estimators=200,learning_rate=0.1,max_depth=5,random_state=10)
gbdt_pipeline = PMMLPipeline([('mapper', mapper), ("classifier", clf)])

sklearn2pmml(gbdt_pipeline,"gbdt.pmml",with_repr=True)

转换为PMML以报告错误。错误如下

Standard output is empty
Standard error:
 28, 2020 4:04:05  org.jpmml.sklearn.Main run
Ϣ: Parsing PKL..
 28, 2020 4:04:05  org.jpmml.sklearn.Main run
Ϣ: Parsed PKL in 80 ms.
 28, 2020 4:04:05  org.jpmml.sklearn.Main run
Ϣ: Converting..
 28, 2020 4:04:05  org.jpmml.sklearn.Main run
: Failed to convert
java.lang.IllegalArgumentException: The value object (Python class sklearn.preprocessing._data.MinMaxScaler) is not a supported Transformer
    at org.jpmml.sklearn.CastFunction.apply(CastFunction.java:43)
    at com.google.common.collect.Lists$TransformingRandomAccessList$1.transform(Lists.java:612)
    at com.google.common.collect.TransformedIterator.next(TransformedIterator.java:47)
    at sklearn_pandas.DataFrameMapper.initializeFeatures(DataFrameMapper.java:72)
    at sklearn.Initializer.encodeFeatures(Initializer.java:41)
    at sklearn.Transformer.updateAndEncodeFeatures(Transformer.java:85)
    at sklearn.pipeline.Pipeline.encodeFeatures(Pipeline.java:83)
    at sklearn2pmml.pipeline.PMMLPipeline.encodePMML(PMMLPipeline.java:203)
    at org.jpmml.sklearn.Main.run(Main.java:145)
    at org.jpmml.sklearn.Main.main(Main.java:94)
Caused by: java.lang.ClassCastException: Cannot cast net.razorvine.pickle.objects.ClassDict to sklearn.Transformer
    at java.lang.Class.cast(Unknown Source)
    at org.jpmml.sklearn.CastFunction.apply(CastFunction.java:41)
    ... 9 more

我使用的scikit-learn是0.22.2，当我使用0.21版本时，没有错误，谢谢你的帮助！

Answer 1

Scikit-Learn在0.21.X和0.22.X开发分支之间重命名了几个模块。例如，sklearn.preprocessing.data模块变成了sklearn.preprocessing._data (前导下划线应该解释为现在是私有模块)。

您使用的是过时的SkLearn2PMML包版本，该版本尚不了解0.22.X模块。请将SkLearn2PMML包更新为最新版本(pip install --upgrade sklearn2pmml)，此问题将得到解决。