mirror of
https://github.com/fenago/data-science.git
synced 2026-05-04 08:31:59 +00:00
fenago
93 lines
3.0 KiB
Plaintext
93 lines
3.0 KiB
Plaintext
import unittest
|
|
import import_ipynb
|
|
import pandas as pd
|
|
import pandas.testing as pd_testing
|
|
import numpy.testing as np_testing
|
|
|
|
from sklearn.ensemble import RandomForestClassifier
|
|
from sklearn.model_selection import train_test_split
|
|
from sklearn.metrics import accuracy_score
|
|
from mlxtend.evaluate import feature_importance_permutation
|
|
|
|
|
|
class Test(unittest.TestCase):
|
|
def setUp(self):
|
|
import Activity9_1
|
|
self.exercises = Activity9_1
|
|
|
|
self.file_url = '../Dataset/KDDCup99.csv'
|
|
self.df = pd.read_csv(self.file_url)
|
|
self.y = self.df.pop('label')
|
|
self.df = pd.get_dummies(self.df)
|
|
|
|
self.X_train, self.X_test, self.y_train, self.y_test = train_test_split(self.df, self.y, test_size=0.3, random_state=1)
|
|
|
|
self.rf_model = RandomForestClassifier(random_state=168)
|
|
self.rf_model.fit(self.X_train, self.y_train)
|
|
|
|
self.train_preds = self.rf_model.predict(self.X_train)
|
|
self.test_preds = self.rf_model.predict(self.X_test)
|
|
|
|
self.train_acc = accuracy_score(self.y_train, self.train_preds)
|
|
self.test_acc = accuracy_score(self.y_test, self.test_preds)
|
|
|
|
self.imp_vals, _ = feature_importance_permutation(predict_method=self.rf_model.predict, X=self.X_train.values,
|
|
y=self.y_train.values, metric='accuracy', num_rounds=1, seed=2)
|
|
|
|
self.perm_varimp_df = pd.DataFrame({'feature': self.X_train.columns, 'importance': self.imp_vals})
|
|
|
|
self.perm_varimp_df.sort_values('importance', ascending=False, inplace=True)
|
|
|
|
self.feature_index = self.df.columns.get_loc("src_bytes")
|
|
|
|
self.class_names = list(sorted(y.unique()))
|
|
|
|
def test_file_url(self):
|
|
self.assertEqual(self.exercises.file_url, self.file_url)
|
|
|
|
def test_df(self):
|
|
pd_testing.assert_frame_equal(self.exercises.df, self.df)
|
|
|
|
def test_y(self):
|
|
np_testing.assert_array_equal(self.exercises.y , self.y)
|
|
|
|
def test_X_train(self):
|
|
np_testing.assert_array_equal(self.exercises.X_train , self.X_train)
|
|
|
|
def test_X_test(self):
|
|
np_testing.assert_array_equal(self.exercises.X_test, self.X_test)
|
|
|
|
def test_y_train(self):
|
|
np_testing.assert_array_equal(self.exercises.y_train , self.y_train)
|
|
|
|
def test_y_test(self):
|
|
np_testing.assert_array_equal(self.exercises.y_test , self.y_test)
|
|
|
|
def test_train_preds(self):
|
|
np_testing.assert_array_equal(self.exercises.train_preds , self.train_preds)
|
|
|
|
def test_test_preds(self):
|
|
np_testing.assert_array_equal(self.exercises.test_preds, self.test_preds)
|
|
|
|
def test_train_acc(self):
|
|
self.assertEqual(self.exercises.train_acc, self.train_acc)
|
|
|
|
def test_test_acc(self):
|
|
self.assertEqual(self.exercises.test_acc, self.test_acc)
|
|
|
|
def test_imp_vals(self):
|
|
np_testing.assert_array_equal(self.exercises.imp_vals, self.imp_vals)
|
|
|
|
def test_perm_varimp_df(self):
|
|
pd_testing.assert_frame_equal(self.exercises.perm_varimp_df, self.perm_varimp_df)
|
|
|
|
def test_feature_index(self):
|
|
self.assertEqual(self.exercises.feature_index, self.feature_index)
|
|
|
|
def test_class_names(self):
|
|
np_testing.assert_array_equal(self.exercises.class_names, self.class_names)
|
|
|
|
|
|
if __name__ == '__main__':
|
|
unittest.main()
|