如何在pyweka上使用attributeselectedclassifier？

Question

我正在将在 weka 上完成的 model 翻译成 python-weka-wrapper3，我不知道如何在 attributeselectedclassifier 上进行评估和搜索选项。

这是 weka 上的 model：

weka.classifiers.meta.AttributeSelectedClassifier -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.GreedyStepwise -B -T -1.7976931348623157E308 -N -1 -num-slots 1" -W weka.classifiers.meta.MultiSearch -- -E FM -search "weka.core.setupgenerator.MathParameter -property classifier.classifier.classifier.numOfBoostingIterations -min 5.0 -max 50.0 -step 1.0 -base 10.0 -expression I" -class-label 1 -algorithm "weka.classifiers.meta.multisearch.DefaultSearch -sample-size 100.0 -initial-folds 2 -subsequent-folds 10 -initial-test-set . -subsequent-test-set . -num-slots 1" -log-file /Applications/weka-3-8-3 -S 1 -W weka.classifiers.meta.Bagging -- -P 100 -S 1 -num-slots 1 -I 100 -W weka.classifiers.meta.FilteredClassifier -- -F "weka.filters.supervised.instance.SMOTE -C 0 -K 3 -P 250.0 -S 1" -S 1 -W weka.classifiers.meta.CostSensitiveClassifier -- -cost-matrix "[0.0 1.0; 1.0 0.0]" -S 1 -W weka.classifiers.trees.ADTree -- -B 10 -E -3 -S 1

我现在有这个：

base = Classifier(classname="weka.classifiers.trees.ADTree", options=["-B", "10", "-E", "-3", "-S", "1"])
cls = SingleClassifierEnhancer(classname="weka.classifiers.meta.CostSensitiveClassifier", 
                                options =["-cost-matrix", "[0.0 1.0; 1.0 0.0]", "-S", "1"])
cls.classifier = base
smote = Filter(classname="weka.filters.supervised.instance.SMOTE", options=["-C", "0", "-K", "3", "-P", "250.0", "-S", "1"])
fc = FilteredClassifier()
fc.filter = smote
fc.classifier = cls
bagging_cls = Classifier(classname="weka.classifiers.meta.Bagging",
                         options=["-P", "100", "-S", "1", "-num-slots", "1", "-I", "100"])
bagging_cls.classifier = fc

multisearch_cls = MultiSearch(
    options = ["-S", "1"])
multisearch_cls.evaluation = "FM"
multisearch_cls.log_file = "/home/pablo/Escritorio/TFG/OUTPUT.txt"
multisearch_cls.search = ["-sample-size", "100", "-initial-folds", "2", "-subsequent-folds", "10",
                          "-initial-test-set", ".", "-subsequent-test-set", ".", "-num-slots", "1"]                        
mparam = MathParameter()
mparam.prop = "numOfBoostingIterations"
mparam.minimum = 5.0
mparam.maximum = 50.0
mparam.step = 1.0
mparam.base = 10.0
mparam.expression = "I"
multisearch_cls.parameters = [mparam]
multisearch_cls.classifier = bagging_cls



AttS_cls = AttributeSelectedClassifier()
AttS_cls.evaluator = "weka.attributeSelection.CfsSubsetEval -P 1 -E 1"
AttS_cls.search = "weka.attributeSelection.GreedyStepwise -B -T -1.7976931348623157E308 -N -1 -num-slots 1"
AttS_cls.classifier = multisearch_cls





train, test = data_modelos_1_2.train_test_split(70.0, Random(1))
AttS_cls.build_classifier(train)
evl = Evaluation(train)
evl.crossvalidate_model(AttS_cls, test, 10, Random(1))

print(AttS_cls)

#graph.plot_dot_graph(AttS_cls.graph)

print("")
print("=== Setup ===")
print("Classifier: " + AttS_cls.to_commandline())
print("Dataset: ")
print(test.relationname)
print("")
print(evl.summary("=== " + str(10) + " -fold Cross-Validation ==="))
print(evl.class_details())

plcls.plot_roc(evl, class_index=[0, 1], wait=True)

但是当我这样做的时候

AttS_cls.evaluator = "weka.attributeSelection.CfsSubsetEval -P 1 -E 1"
AttS_cls.search = "weka.attributeSelection.GreedyStepwise -B -T -1.7976931348623157E308 -N -1 -num-slots 1"

它给我这个错误：

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
/tmp/ipykernel_40724/2750622902.py in <module>
     30 
     31 AttS_cls = AttributeSelectedClassifier()
---> 32 AttS_cls.search = "weka.attributeSelection.GreedyStepwise"
     33 AttS_cls.classifier = multisearch_cls
     34 

/usr/local/lib/python3.8/dist-packages/weka/classifiers.py in search(self, search)
    435         :type search: ASSearch
    436         """
--> 437         javabridge.call(self.jobject, "setSearch", "(Lweka/attributeSelection/ASSearch;)V", search.jobject)
    438 
    439 

AttributeError: 'str' object has no attribute 'jobject'

我知道我必须将它们设置为对象，因为它会引发此错误，因为我尝试将它们设置为字符串，但我不知道如何设置。

Answer 1

您需要实例化ASSearch和ASEvaluation对象。 如果你有命令行，你可以像这样使用from_commandline辅助方法：

from weka.core.classes import from_commandline, get_classname
from weka.attribute_selection import ASSearch
from weka.attribute_selection import ASEvaluation

search = from_commandline('weka.attributeSelection.GreedyStepwise -B -T -1.7976931348623157E308 -N -1 -num-slots 1', classname=get_classname(ASSearch))
evaluation = from_commandline('weka.attributeSelection.CfsSubsetEval -P 1 -E 1', classname=get_classname(ASEvaluation))

from_commandline方法的第二个参数是您要使用而不是classname的包装器的类OptionHandler 。 为简单起见，我导入了正确的包装器，然后使用get_classname方法返回包装器 class 的点符号。这样我就可以避免类名字符串中的意外拼写错误。 此外，通过使用单引号，您不必担心命令行中的 Weka 引号，您可以直接使用 Weka 命令行字符串。

您还可以使用相同的方法来实例化AttributeSelectedClassifier包装器本身，而不必 go 通过分别实例化搜索和评估：

from weka.core.classes import from_commandline, get_classname
from weka.classifiers import AttributeSelectedClassifier

cls = from_commandline('weka.classifiers.meta.AttributeSelectedClassifier -E "weka.attributeSelection.CfsSubsetEval -P 1 -E 1" -S "weka.attributeSelection.GreedyStepwise -B -T -1.7976931348623157E308 -N -1 -num-slots 1" -W weka.classifiers.meta.MultiSearch -- -E FM -search "weka.core.setupgenerator.MathParameter -property classifier.classifier.classifier.numOfBoostingIterations -min 5.0 -max 50.0 -step 1.0 -base 10.0 -expression I" -class-label 1 -algorithm "weka.classifiers.meta.multisearch.DefaultSearch -sample-size 100.0 -initial-folds 2 -subsequent-folds 10 -initial-test-set . -subsequent-test-set . -num-slots 1" -log-file /Applications/weka-3-8-3 -S 1 -W weka.classifiers.meta.Bagging -- -P 100 -S 1 -num-slots 1 -I 100 -W weka.classifiers.meta.FilteredClassifier -- -F "weka.filters.supervised.instance.SMOTE -C 0 -K 3 -P 250.0 -S 1" -S 1 -W weka.classifiers.meta.CostSensitiveClassifier -- -cost-matrix "[0.0 1.0; 1.0 0.0]" -S 1 -W weka.classifiers.trees.ADTree -- -B 10 -E -3 -S 1', get_classname(AttributeSelectedClassifier))