def multinomial(mi): MessageFeatures.test_fold = -1 mn = Multinomial(mi) mn.train() mn.test() #print(mn.correct) pass
def kfcvm(mi, k=10): correct = [] tested = [] tot = 0 cor = 0 for i in range(k): #mark_test_set(mi, k, i) MessageFeatures.test_fold = i MessageFeatures.folds = k mn = Multinomial(mi) mn.train() mn.test_marked()
from LogisticRegression import LogisticRegression
from Multinomial import Multinomial
def loadData(file):
data = pd.read_csv(file, header=None).fillna(0)
return data.drop_duplicates()
if __name__ == '__main__':
print("1 Spambase Logistic Regression")
LogisticRegression(loadData('./data/spambase.csv')).validate()
print("1 Breast Cancer Logistic Regression")
LogisticRegression(loadData('./data/breastcancer.csv')).validate()
print("1 Diabetes Logistic Regression")
LogisticRegression(loadData('./data/diabetes.csv')).validate()
print("2 Multivariate Bernoulli")
Multinomial(True).validate('./data/20NG_data/train_data.csv',
'./data/20NG_data/train_label.csv',
'./data/20NG_data/test_data.csv',
'./data/20NG_data/test_label.csv')
print("2 Multinomial")
Multinomial().validate('./data/20NG_data/train_data.csv',
'./data/20NG_data/train_label.csv',
'./data/20NG_data/test_data.csv',
'./data/20NG_data/test_label.csv')
dataSet = dataSet.drop_duplicates()
return dataSet
if __name__ == '__main__':
spambaseFileLocation = 'spambase.csv'
spambaseDataSet = importData(spambaseFileLocation)
breastCancerFileLocation = 'breastcancer.csv'
breastCancerDataSet = importData(breastCancerFileLocation)
diabetesFileLocation = 'diabetes.csv'
diabetesDataSet = importData(diabetesFileLocation)
print('Spambase Dataset - Logistic Regression')
spambaseLogisticRegression = LogisticRegression(spambaseDataSet, 0.75,
0.00001)
spambaseLogisticRegression.validate()
print('Breast Cancer Dataset - Logistic Regression')
breastCancerLogisticRegression = LogisticRegression(
breastCancerDataSet, 0.75, 0.00001)
breastCancerLogisticRegression.validate()
print('Pima Indian Diabetes Dataset - Logistic Regression')
diabetesLogisticRegression = LogisticRegression(diabetesDataSet, 0.1,
0.0000001)
diabetesLogisticRegression.validate()
multivariateBernoulli = MultivariateBernoulli()
multivariateBernoulli.run('train.data', 'train.label', 'test.data',
'test.label')
multinomial = Multinomial()
multinomial.run('train.data', 'train.label', 'test.data', 'test.label')
