def test_dense_toy_dataset_model_with_missing_labels():
model = dense_model(toy_dataset())
x_train, y_train, x_test, y_test = toy_dataset()
with pytest.raises(AssertionError) as err:
plot_loss(model, ((x_train, y_train), (x_test, y_test)),
dataset_labels=('tranining set', ))
assert err.type is AssertionError
assert "Datasets and labels length must be the same" in str(err.value)
def test_to_model_params_dense_net_values_are_close_to_original(self):
model = dense_model(toy_dataset())
sut, model_weights, is_bias, biases = self.get_sut_and_test_input(
model)
input = self.weights_as_single_vector(model_weights)
recreated_params = sut.to_model_params(input)
recreated_params = self.unwrap_list(recreated_params)
for weights_matrix, is_b, bias, recreated_weights in zip(
model_weights, is_bias, biases, recreated_params):
self.is_close(recreated_weights, is_b, bias, weights_matrix)
def test_dense_toy_dataset_model():
model = dense_model(toy_dataset())
_, _, x_test, y_test = toy_dataset()
plot_loss(model, (x_test, y_test))
def test_dense_toy_dataset_number_of_points_41():
model = dense_model(toy_dataset())
_, _, x_test, y_test = toy_dataset()
plot_loss(model, (x_test, y_test), number_of_points=41)
def test_dense_toy_dataset_model_with_labels():
model = dense_model(toy_dataset())
x_train, y_train, x_test, y_test = toy_dataset()
plot_loss(model, ((x_train, y_train), (x_test, y_test)),
dataset_labels=('tranining set', 'test set'))
def test_dense_toy_dataset_model_multiple_datasets():
model = dense_model(toy_dataset())
x_train, y_train, x_test, y_test = toy_dataset()
plot_loss(model, ((x_train, y_train), (x_test, y_test)))
def test_dense_toy_dataset_model():
model = dense_model(toy_dataset())
_, _, x_test, y_test = toy_dataset()
plot_loss_3D(model, "levels", x_test, y_test)
def test_dense_toy_dataset_num_points_3():
model = dense_model(toy_dataset())
_, _, x_test, y_test = toy_dataset()
plot_loss_3D(model, "levels", x_test, y_test, number_of_points=3)
def setup(cls):
if not hasattr(
cls, 'initialized'): # avoid learning networks multiple times
cls.initialized = True
cls.dense = dense_model(toy_dataset())
cls.conv = conv_model(mnist_dataset())