def setUpClass(cls):
df = get_dataset("infert").as_df()
# remove : and ' ' from column names, and encode categorical column
df.columns = [i.replace(': ', '') for i in df.columns]
assert is_string_dtype(df['education_str'].dtype)
df = (OneHotVectorizer() << ['education_str']).fit_transform(df)
assert 'education_str' not in df.columns
cls.X, cls.y = split_features_and_label(df, 'case')
def get_iris():
df = get_dataset("iris").as_df()
df.drop(['Label'], inplace=True, axis=1)
df['Label'] = df['Species']
df.drop(['Species'], inplace=True, axis=1)
df.drop(['Setosa'], inplace=True, axis=1)
X, y = split_features_and_label(df, 'Label')
return X, y
def infert_df(label_name):
df = get_dataset('infert').as_df()
df = (OneHotVectorizer() << 'education_str').fit_transform(df)
X, y = split_features_and_label(df, label_name)
return X, y
from nimbusml.decomposition import FactorizationMachineBinaryClassifier
from nimbusml.ensemble import LightGbmClassifier
from nimbusml.linear_model import FastLinearClassifier
from nimbusml.linear_model import LogisticRegressionBinaryClassifier
from nimbusml.linear_model import LogisticRegressionClassifier
from nimbusml.naive_bayes import NaiveBayesClassifier
from nimbusml.tests.test_utils import split_features_and_label
from sklearn.model_selection import train_test_split
from sklearn.utils.testing import assert_almost_equal, assert_equal
# use iris dataset
np.random.seed(0)
df = get_dataset("iris").as_df()
df.drop(['Species'], inplace=True, axis=1)
df.Label = [1 if x == 1 else 0 for x in df.Label]
features, labels = split_features_and_label(df, 'Label')
X_train, X_test, y_train, y_test = \
train_test_split(features, labels)
# 3 class dataset with string labels
np.random.seed(0)
df = get_dataset("iris").as_df()
df.drop(['Species'], inplace=True, axis=1)
_str_map = {0: 'Red', 1: 'Green', 2: 'Blue'}
df.Label = df.Label.apply(lambda x: _str_map[x])
features_3class, labels_3class = split_features_and_label(df, 'Label')
X_train_3class, X_test_3class, y_train_3class, y_test_3class = \
train_test_split(features_3class, labels_3class)
# fit classifier, return sum of probabilites
from nimbusml.linear_model import FastLinearBinaryClassifier, \
AveragedPerceptronBinaryClassifier
from nimbusml.linear_model import LogisticRegressionBinaryClassifier, \
SgdBinaryClassifier
# from nimbusml.linear_model import SymSgdBinaryClassifier
from nimbusml.multiclass import OneVsRestClassifier
from nimbusml.tests.test_utils import split_features_and_label
from sklearn.model_selection import train_test_split
from sklearn.utils.testing import assert_equal, assert_not_equal, \
assert_greater
# use iris dataset
np.random.seed(0)
df = get_dataset("iris").as_df()
df.drop(['Species'], inplace=True, axis=1)
features, labels = split_features_and_label(df, 'Label')
X_train, X_test, y_train, y_test = \
train_test_split(features, labels)
# fit classifier, return sum of probabilites
def proba_average(ovr):
ovr.fit(X_train, y_train)
return ovr.predict_proba(X_test).sum(axis=1).mean()
# fit classifier, return sum of decision values
def decfun_average(ovr):
ovr.fit(X_train, y_train)
return ovr.decision_function(X_test).sum(axis=1).mean()
def setUpClass(cls):
df = get_dataset("iris").as_df()
df.drop(['Species'], inplace=True, axis=1)
cls.X, cls.y = split_features_and_label(df, 'Label')
