from __future__ import absolute_import, print_function, division, unicode_literals
import unittest
from xcessiv import models
from sklearn.ensemble import RandomForestClassifier
class TestReturnTrainDataFromJSON(unittest.TestCase):
def setUp(self):
self.extraction = models.Extraction()
self.extraction.main_dataset['source'] = ''.join([
"from sklearn.datasets import load_digits\n",
"\n",
"\n",
"def extract_main_dataset():\n",
" X, y = load_digits(return_X_y=True)\n",
" return X, y"
])
def test_main_is_train(self):
X, y = self.extraction.return_train_dataset()
assert X.shape == (1797, 64)
assert y.shape == (1797,)
def test_split_main_for_test(self):
self.extraction.test_dataset['method'] = 'split_from_main'
self.extraction.test_dataset['split_ratio'] = 0.1
self.extraction.test_dataset['split_seed'] = 8
X, y = self.extraction.return_train_dataset()
assert X.shape == (1617, 64)
assert y.shape == (1617,)
class TestReturnTestDataFromJSON(unittest.TestCase):
def setUp(self):
self.extraction = models.Extraction()
self.extraction.main_dataset['source'] = ''.join([
"from sklearn.datasets import load_digits\n",
"\n",
"\n",
"def extract_main_dataset():\n",
" X, y = load_digits(return_X_y=True)\n",
" return X, y"
])
self.extraction.test_dataset['method'] = 'split_from_main'
self.extraction.test_dataset['split_ratio'] = 0.1
self.extraction.test_dataset['split_seed'] = 8
def test_split_main_for_test(self):
X, y = self.extraction.return_test_dataset()
assert X.shape == (180, 64)
assert y.shape == (180,)
def test_test_dataset_from_source(self):
self.extraction.test_dataset["method"] = "source"
self.extraction.test_dataset["source"] = ''.join([
"from sklearn.datasets import load_digits\n",
"def extract_test_dataset():\n",
" X, y = load_digits(return_X_y=True)\n",
" return X, y"
])
X, y = self.extraction.return_test_dataset()
assert X.shape == (1797, 64)
assert y.shape == (1797,)
class TestReturnEstimator(unittest.TestCase):
def setUp(self):
self.base_learner_origin = models.BaseLearnerOrigin(
source=''.join([
"from sklearn.ensemble import RandomForestClassifier\n",
"base_learner = RandomForestClassifier(random_state=8)"
])
)
def test_return_estimator_from_json(self):
est = self.base_learner_origin.return_estimator()
assert isinstance(est, RandomForestClassifier)
