def J48(data, rnm):
data.class_is_last()
fc = FilteredClassifier()
fc.classifier = Classifier(classname="weka.classifiers.trees.J48", options=["-C", "0.25", "-M", "2"])
fc.filter = Filter(classname="weka.filters.unsupervised.attribute.Remove", options=["-R", "first"])
pred_output = PredictionOutput(classname="weka.classifiers.evaluation.output.prediction.CSV", options=["-p", "1"])
folds = 10
evl = Evaluation(data)
evl.crossvalidate_model(fc, data, folds, Random(1), pred_output)
fc.build_classifier(data)
f0 = open(rnm + '_J48_Tree.txt', 'w')
print >> f0, "Filename: ", rnm
print >> f0, '\n\n'
print >> f0, str(fc)
f0.close()
f1 = open(rnm + '_J48_Prediction.txt', 'w')
print >> f1, 'Filename:', rnm
print >> f1, 'Prediction Summary:', (pred_output.buffer_content())
f1.close()
f2 = open(rnm + '_J48_Evaluation.txt', 'w')
print >> f2, 'Filename:', rnm
print >> f2, 'Evaluation Summary:', (evl.summary())
print >> f2, '\n\n\n'
print >> f2, (evl.class_details())
f2.close()
plot_roc(evl, class_index=[0,1], title=rnm, key_loc='best', outfile=rnm + '_J48_ROC.png', wait=False)
value_J48 = str(evl.percent_correct)
return value_J48
print(group)
train = data_dir + os.sep + group + "_Cal.arff"
test = data_dir + os.sep + group + "_Test.arff"
pred = data_dir + os.sep + group + "_Val.arff"
loader = Loader(classname="weka.core.converters.ArffLoader")
print(train)
train_data = loader.load_file(train)
train_data.class_index = train_data.attribute_by_name(
"reference value").index
print(test)
test_data = loader.load_file(test)
test_data.class_index = test_data.attribute_by_name(
"reference value").index
print(pred)
pred_data = loader.load_file(pred)
pred_data.class_index = pred_data.attribute_by_name(
"reference value").index
cls = FilteredClassifier()
cls.classifier = Classifier(
classname="weka.classifiers.functions.LinearRegression",
options=["-S", "1", "-C"])
cls.filter = Filter(classname="weka.filters.unsupervised.attribute.Remove",
options=["-R", "first"])
cls.build_classifier(train_data)
evl = Evaluation(train_data)
evl.test_model(cls, test_data)
print(evl.summary())
jvm.stop()
fname = data_dir + os.sep + "ReutersGrain-test.arff"
print("\nLoading dataset: " + fname + "\n")
loader = Loader(classname="weka.core.converters.ArffLoader")
test = loader.load_file(fname)
test.set_class_index(test.num_attributes() - 1)
setups = (
("weka.classifiers.trees.J48", []),
("weka.classifiers.bayes.NaiveBayes", []),
("weka.classifiers.bayes.NaiveBayesMultinomial", []),
("weka.classifiers.bayes.NaiveBayesMultinomial", ["-C"]),
("weka.classifiers.bayes.NaiveBayesMultinomial", ["-C", "-L", "-S"])
)
# cross-validate classifiers
for setup in setups:
classifier, opt = setup
print("\n--> %s (filter options: %s)\n" % (classifier, " ".join(opt)))
cls = FilteredClassifier()
cls.set_classifier(Classifier(classname=classifier))
cls.set_filter(Filter(classname="weka.filters.unsupervised.attribute.StringToWordVector", options=opt))
cls.build_classifier(data)
evl = Evaluation(test)
evl.test_model(cls, test)
print("Accuracy: %0.0f%%" % evl.percent_correct())
tcdata = plc.generate_thresholdcurve_data(evl, 0)
print("AUC: %0.3f" % plc.get_auc(tcdata))
print(evl.to_matrix("Matrix:"))
jvm.stop()
print("Train/test/predict...")
groups = ["DataSet1", "DataSet2"]
# groups = ["DataSet2"]
for group in groups:
print(group)
train = data_dir + os.sep + group + "_Cal.arff"
test = data_dir + os.sep + group + "_Test.arff"
pred = data_dir + os.sep + group + "_Val.arff"
loader = Loader(classname="weka.core.converters.ArffLoader")
print(train)
train_data = loader.load_file(train)
train_data.class_index = train_data.attribute_by_name("reference value").index
print(test)
test_data = loader.load_file(test)
test_data.class_index = test_data.attribute_by_name("reference value").index
print(pred)
pred_data = loader.load_file(pred)
pred_data.class_index = pred_data.attribute_by_name("reference value").index
cls = FilteredClassifier()
cls.classifier = Classifier(classname="weka.classifiers.functions.LinearRegression", options=["-S", "1", "-C"])
cls.filter = Filter(classname="weka.filters.unsupervised.attribute.Remove", options=["-R", "first"])
cls.build_classifier(train_data)
evl = Evaluation(train_data)
evl.test_model(cls, test_data)
print(evl.summary())
jvm.stop()
jvm.start(system_cp=True, packages=True, max_heap_size="512m")
# Train classifier
loader = Loader(classname="weka.core.converters.ArffLoader", options=["-charset", "UTF-8"])
train_data = loader.load_file(os.path.dirname(os.path.realpath(__file__)) + "/datasets/train.arff")
train_data.class_is_last()
string_to_word_vector_filter = Filter(classname="weka.filters.unsupervised.attribute.StringToWordVector")
cls = Classifier(classname="weka.classifiers.bayes.NaiveBayesMultinomial")
fc = FilteredClassifier()
fc.filter = string_to_word_vector_filter
fc.classifier = cls
fc.build_classifier(train_data)
# Create test data
class_att = Attribute.create_nominal("class", ["good", "neutral", "bad"])
str_att = Attribute.create_string("title")
test_dataset = Instances.create_instances(
name="test_news_set",
atts=[str_att, class_att],
capacity=1
)
inst = Instance.create_instance([Instance.missing_value(), Instance.missing_value()])
test_dataset.add_instance(inst)
test_dataset.get_instance(0).set_string_value(0, article['processed']['title'])
fname = data_dir + os.sep + "ReutersGrain-test.arff"
print("\nLoading dataset: " + fname + "\n")
loader = Loader(classname="weka.core.converters.ArffLoader")
test = loader.load_file(fname)
test.class_is_last()
setups = (
("weka.classifiers.trees.J48", []),
("weka.classifiers.bayes.NaiveBayes", []),
("weka.classifiers.bayes.NaiveBayesMultinomial", []),
("weka.classifiers.bayes.NaiveBayesMultinomial", ["-C"]),
("weka.classifiers.bayes.NaiveBayesMultinomial", ["-C", "-L", "-stopwords-handler", "weka.core.stopwords.Rainbow"])
)
# cross-validate classifiers
for setup in setups:
classifier, opt = setup
print("\n--> %s (filter options: %s)\n" % (classifier, " ".join(opt)))
cls = FilteredClassifier()
cls.classifier = Classifier(classname=classifier)
cls.filter = Filter(classname="weka.filters.unsupervised.attribute.StringToWordVector", options=opt)
cls.build_classifier(data)
evl = Evaluation(test)
evl.test_model(cls, test)
print("Accuracy: %0.0f%%" % evl.percent_correct)
tcdata = plc.generate_thresholdcurve_data(evl, 0)
print("AUC: %0.3f" % plc.get_auc(tcdata))
print(evl.matrix("Matrix:"))
jvm.stop()
evl.crossvalidate_model(fc, train, 5, Random(1))
# Prints Out Confusion Matrix along with other summary statistics
print("LMT (SMOTE balanced classes) CV = 5 Error: %.2f%%" %
(evl.percent_incorrect))
print(evl.matrix()) #Confusion Matrix
# Plots ROC
plcls.plot_roc(evl, class_index=[0, 1], wait=True)
# Extra Summary
print(evl.summary())
print(evl.class_details())
# Evaluate the classifier on test set
fc.build_classifier(train)
tevl = Evaluation(test)
tevl.test_model(fc, test)
# Prints Out Confusion Matrix along with other summary statistics
print("LMT (SMOTE balanced classes) Test Error: %.2f%%" %
(tevl.percent_incorrect))
print(tevl.matrix()) #Confusion Matrix
# Plots ROC
plcls.plot_roc(tevl, class_index=[0, 1], wait=True)
# Extra Summary
print(tevl.summary())
print(tevl.class_details())
