def test_accuracy_string_index(self):
sample_ids = ["V01", "V02", "V03", "V04"]
results = pd.Series(["a", "b", "c", "a"], index=sample_ids)
labels = pd.Series(["a", "b", "c", "b"], index=sample_ids)
dataset = self.create_dataset(labels, sample_ids=sample_ids)
classified = ClassifiedDataSet(dataset, results)
self.assertAlmostEqual(classified.compute_accuracy(), 0.75, places=2)
def test_accuracy_string_index(self):
sample_ids = ["V01", "V02", "V03", "V04"]
results = pd.Series(["a", "b", "c", "a"], index=sample_ids)
labels = pd.Series(["a", "b", "c", "b"], index=sample_ids)
dataset = self.create_dataset(labels, sample_ids=sample_ids)
classified = ClassifiedDataSet(dataset, results)
self.assertAlmostEqual(classified.compute_accuracy(), 0.75, places=2)
def test_accuracy_unequal_lengths(self):
results = pd.Series(["a", "b", "c", "a"],
index=["V01", "V02", "V03", "V04"])
labels = pd.Series(["a", "b", "c"], index=["V01", "V02", "V03"])
dataset = self.create_dataset(labels=labels)
classified = ClassifiedDataSet(dataset, results)
self.assertRaises(ValueError, classified.compute_accuracy)
def test_accuracy_integer_index(self):
dataset = self.create_dataset(labels=["a", "b", "c", "b"])
classified = ClassifiedDataSet(dataset,
pd.Series(["a", "b", "c", "a"]))
self.assertAlmostEqual(classified.compute_accuracy(), 0.75, places=2)
def test_accuracy_dataset_unlabelled(self):
results = pd.Series(["a", "b", "c", "a"],
index=["V01", "V02", "V03", "V04"])
dataset = DataSet([[1, 2, 3], [4, 5, 6], [7, 8, 9], [10, 11, 12]])
classified = ClassifiedDataSet(dataset, results)
self.assertRaises(UnlabelledDataSetError, classified.compute_accuracy)
def test_accuracy_integer_index(self):
dataset = self.create_dataset(labels=["a", "b", "c", "b"])
classified = ClassifiedDataSet(dataset, pd.Series(["a", "b", "c",
"a"]))
self.assertAlmostEqual(classified.compute_accuracy(), 0.75, places=2)
