def test_list_of_tensors(self):
range20 = np.expand_dims(np.arange(20), 1)
dataset = TensorDataset((range20, range20 * 2), range20 * 3)
self.assertEqual(len(dataset), 20)
self.assertEqual(type(dataset[0]), tuple)
self.assertEqual(type(dataset[0][0]), tuple)
self.assertEqual(type(dataset[0][-1]), np.ndarray)
for i in range(20):
self.assertEqual(dataset[i][0][0], i)
self.assertEqual(dataset[i][0][1], i * 2)
self.assertEqual(dataset[i][1], i * 3)
dataset = TensorDataset((range20, range20 * 2), (range20 * 3, range20 * 4))
self.assertEqual(len(dataset), 20)
self.assertEqual(type(dataset[0]), tuple)
self.assertEqual(type(dataset[1]), tuple)
self.assertEqual(type(dataset[0][0]), tuple)
self.assertEqual(type(dataset[0][1]), tuple)
for i in range(20):
self.assertEqual(type(dataset[i][0][0]), np.ndarray)
self.assertEqual(type(dataset[i][0][1]), np.ndarray)
self.assertEqual(dataset[i][0][0], i)
self.assertEqual(dataset[i][0][1], i * 2)
self.assertEqual(type(dataset[i][1][0]), np.ndarray)
self.assertEqual(type(dataset[i][1][1]), np.ndarray)
self.assertEqual(dataset[i][1][0], i * 3)
self.assertEqual(dataset[i][1][1], i * 4)
def test_fitting_generator_calls_with_longer_validation_set(self):
train_real_steps_per_epoch = 30
train_batch_size = ModelTest.batch_size
train_final_batch_missing_samples = 7
train_size = train_real_steps_per_epoch * train_batch_size - \
train_final_batch_missing_samples
train_x = torch.rand(train_size, 1)
train_y = torch.rand(train_size, 1)
train_dataset = TensorDataset(train_x, train_y)
train_generator = DataLoader(train_dataset, train_batch_size)
valid_real_steps_per_epoch = 40
valid_batch_size = 15
valid_final_batch_missing_samples = 3
valid_size = valid_real_steps_per_epoch * valid_batch_size - \
valid_final_batch_missing_samples
valid_x = torch.rand(valid_size, 1)
valid_y = torch.rand(valid_size, 1)
valid_dataset = TensorDataset(valid_x, valid_y)
valid_generator = DataLoader(valid_dataset, valid_batch_size)
mock_train_generator = IterableMock(train_generator)
mock_valid_generator = IterableMock(valid_generator)
self.model.fit_generator(mock_train_generator,
mock_valid_generator,
epochs=ModelTest.epochs)
expected_train_calls = ['__len__'] + \
(['__iter__'] + ['__next__'] * train_real_steps_per_epoch) * ModelTest.epochs
expected_valid_calls = ['__len__'] + \
(['__iter__'] + ['__next__'] * valid_real_steps_per_epoch) * ModelTest.epochs
self.assertEqual(mock_train_generator.calls, expected_train_calls)
self.assertEqual(mock_valid_generator.calls, expected_valid_calls)
def test_fitting_with_data_loader(self):
train_real_steps_per_epoch = 30
train_batch_size = ModelTest.batch_size
train_final_batch_missing_samples = 7
train_size = train_real_steps_per_epoch * train_batch_size - \
train_final_batch_missing_samples
train_x = torch.rand(train_size, 1)
train_y = torch.rand(train_size, 1)
train_dataset = TensorDataset(train_x, train_y)
train_generator = DataLoader(train_dataset, train_batch_size)
valid_real_steps_per_epoch = 10
valid_batch_size = 15
valid_final_batch_missing_samples = 3
valid_size = valid_real_steps_per_epoch * valid_batch_size - \
valid_final_batch_missing_samples
valid_x = torch.rand(valid_size, 1)
valid_y = torch.rand(valid_size, 1)
valid_dataset = TensorDataset(valid_x, valid_y)
valid_generator = DataLoader(valid_dataset, valid_batch_size)
logs = self.model.fit_generator(train_generator,
valid_generator,
epochs=ModelTest.epochs,
steps_per_epoch=None,
validation_steps=None,
callbacks=[self.mock_callback])
params = {
'epochs': ModelTest.epochs,
'steps': train_real_steps_per_epoch,
'valid_steps': valid_real_steps_per_epoch
}
self._test_callbacks_train(params, logs)
def test_multiple_tensors(self):
range20 = np.expand_dims(np.arange(20), 1)
dataset = TensorDataset(range20, range20 * 2, range20 * 3)
self.assertEqual(len(dataset), 20)
self.assertEqual(type(dataset[0]), tuple)
for i in range(20):
self.assertEqual(dataset[i][0], i)
self.assertEqual(dataset[i][1], i * 2)
self.assertEqual(dataset[i][2], i * 3)
def test_evaluate_data_loader_with_progress_bar_coloring(self):
x = torch.rand(ModelFittingTestCase.evaluate_dataset_len, 1)
y = torch.rand(ModelFittingTestCase.evaluate_dataset_len, 1)
dataset = TensorDataset(x, y)
generator = DataLoader(dataset, ModelFittingTestCase.batch_size)
_, _ = self.model.evaluate_generator(generator, verbose=True)
self.assertStdoutContains(["%", "[32m", "[35m", "[36m", "[94m", "\u2588"])
def test_evaluate_data_loader_multi_input(self):
x1 = torch.rand(ModelMultiInputTest.evaluate_dataset_len, 1)
x2 = torch.rand(ModelMultiInputTest.evaluate_dataset_len, 1)
y = torch.rand(ModelMultiInputTest.evaluate_dataset_len, 1)
dataset = TensorDataset((x1, x2), y)
generator = DataLoader(dataset, ModelMultiInputTest.batch_size)
loss, pred_y = self.model.evaluate_generator(generator,
return_pred=True)
self.assertEqual(type(loss), float)
self.assertEqual(pred_y.shape,
(ModelMultiInputTest.evaluate_dataset_len, 1))
def test_evaluate_data_loader(self):
x = torch.rand(ModelTest.evaluate_dataset_len, 1)
y = torch.rand(ModelTest.evaluate_dataset_len, 1)
dataset = TensorDataset(x, y)
generator = DataLoader(dataset, ModelTest.batch_size)
loss, metrics, pred_y = self.model.evaluate_generator(generator,
return_pred=True)
self.assertEqual(type(loss), float)
self.assertEqual(type(metrics), np.ndarray)
self.assertEqual(metrics.tolist(),
self.batch_metrics_values + self.epoch_metrics_values)
self.assertEqual(pred_y.shape, (ModelTest.evaluate_dataset_len, 1))
def test_evaluate_data_loader_multi_output(self):
x = torch.rand(ModelMultiOutputTest.evaluate_dataset_len, 1)
y1 = torch.rand(ModelMultiOutputTest.evaluate_dataset_len, 1)
y2 = torch.rand(ModelMultiOutputTest.evaluate_dataset_len, 1)
dataset = TensorDataset(x, (y1, y2))
generator = DataLoader(dataset, ModelMultiOutputTest.batch_size)
loss, pred_y = self.model.evaluate_generator(generator,
return_pred=True)
self.assertEqual(type(loss), float)
for pred in pred_y:
self.assertEqual(type(pred), np.ndarray)
self.assertEqual(pred.shape,
(ModelMultiOutputTest.evaluate_dataset_len, 1))
valid_y = np.random.randint(num_classes,
size=num_valid_samples).astype('int64')
num_test_samples = 200
test_x = np.random.randn(num_test_samples, num_features).astype('float32')
test_y = np.random.randint(num_classes, size=num_test_samples).astype('int64')
cuda_device = 0
device = torch.device("cuda:%d" %
cuda_device if torch.cuda.is_available() else "cpu")
# Define the network
network = nn.Sequential(nn.Linear(num_features, hidden_state_size), nn.ReLU(),
nn.Linear(hidden_state_size, num_classes))
# We need to use dataloaders (i.e. an iterable of batches) with Experiment
train_loader = DataLoader(TensorDataset(train_x, train_y), batch_size=32)
valid_loader = DataLoader(TensorDataset(valid_x, valid_y), batch_size=32)
test_loader = DataLoader(TensorDataset(test_x, test_y), batch_size=32)
# Everything is saved in ./expt/my_classification_network
expt = Experiment('./expt/my_classification_network',
network,
device=device,
optimizer='sgd',
task='classif')
expt.train(train_loader, valid_loader, epochs=5)
expt.test(test_loader)
def _get_data_loader(self, *tensors):
return DataLoader(TensorDataset(*tensors),
batch_size=SKLearnMetricsTest.batch_size)
def test_one_tensor(self):
range20 = np.expand_dims(np.arange(20), 1)
dataset = TensorDataset(range20)
self.assertEqual(len(dataset), 20)
for i in range(20):
self.assertEqual(dataset[i], np.array([i]))