PEP8 fixes in tests.
This commit is contained in:
parent
309f586424
commit
c10945f53a
@ -108,8 +108,8 @@ def test_stacked_lstm_char_prediction():
|
|||||||
y = np.zeros((len(sentences), number_of_chars), dtype=np.bool)
|
y = np.zeros((len(sentences), number_of_chars), dtype=np.bool)
|
||||||
for i, sentence in enumerate(sentences):
|
for i, sentence in enumerate(sentences):
|
||||||
for t, char in enumerate(sentence):
|
for t, char in enumerate(sentence):
|
||||||
X[i, t, ord(char)-ord('a')] = 1
|
X[i, t, ord(char) - ord('a')] = 1
|
||||||
y[i, ord(next_chars[i])-ord('a')] = 1
|
y[i, ord(next_chars[i]) - ord('a')] = 1
|
||||||
|
|
||||||
# learn the alphabet with stacked LSTM
|
# learn the alphabet with stacked LSTM
|
||||||
model = Sequential([
|
model = Sequential([
|
||||||
@ -123,7 +123,7 @@ def test_stacked_lstm_char_prediction():
|
|||||||
# prime the model with 'ab' sequence and let it generate the learned alphabet
|
# prime the model with 'ab' sequence and let it generate the learned alphabet
|
||||||
sentence = alphabet[:sequence_length]
|
sentence = alphabet[:sequence_length]
|
||||||
generated = sentence
|
generated = sentence
|
||||||
for iteration in range(number_of_chars-sequence_length):
|
for iteration in range(number_of_chars - sequence_length):
|
||||||
x = np.zeros((1, sequence_length, number_of_chars))
|
x = np.zeros((1, sequence_length, number_of_chars))
|
||||||
for t, char in enumerate(sentence):
|
for t, char in enumerate(sentence):
|
||||||
x[0, t, ord(char) - ord('a')] = 1.
|
x[0, t, ord(char) - ord('a')] = 1.
|
||||||
|
@ -790,7 +790,7 @@ class TestBackend(object):
|
|||||||
|
|
||||||
# len max_time_steps array of batch_size x depth matrices
|
# len max_time_steps array of batch_size x depth matrices
|
||||||
inputs = ([input_prob_matrix_0[t, :][np.newaxis, :]
|
inputs = ([input_prob_matrix_0[t, :][np.newaxis, :]
|
||||||
for t in range(seq_len_0)] + # Pad to max_time_steps = 8
|
for t in range(seq_len_0)] + # Pad to max_time_steps = 8
|
||||||
2 * [np.zeros((1, depth), dtype=np.float32)])
|
2 * [np.zeros((1, depth), dtype=np.float32)])
|
||||||
|
|
||||||
inputs = KTF.variable(np.asarray(inputs).transpose((1, 0, 2)))
|
inputs = KTF.variable(np.asarray(inputs).transpose((1, 0, 2)))
|
||||||
@ -899,7 +899,7 @@ class TestBackend(object):
|
|||||||
def test_foldl(self):
|
def test_foldl(self):
|
||||||
x = np.random.rand(10, 3).astype(np.float32)
|
x = np.random.rand(10, 3).astype(np.float32)
|
||||||
for K in [KTF, KTH]:
|
for K in [KTF, KTH]:
|
||||||
kx = K.eval(K.foldl(lambda a, b: a+b, x))
|
kx = K.eval(K.foldl(lambda a, b: a + b, x))
|
||||||
|
|
||||||
assert (3,) == kx.shape
|
assert (3,) == kx.shape
|
||||||
assert_allclose(x.sum(axis=0), kx, atol=1e-05)
|
assert_allclose(x.sum(axis=0), kx, atol=1e-05)
|
||||||
@ -911,8 +911,8 @@ class TestBackend(object):
|
|||||||
# right to left we have no such problem and the result is larger
|
# right to left we have no such problem and the result is larger
|
||||||
x = np.array([1e-20, 1e-20, 10, 10, 10], dtype=np.float32)
|
x = np.array([1e-20, 1e-20, 10, 10, 10], dtype=np.float32)
|
||||||
for K in [KTF, KTH]:
|
for K in [KTF, KTH]:
|
||||||
p1 = K.eval(K.foldl(lambda a, b: a*b, x))
|
p1 = K.eval(K.foldl(lambda a, b: a * b, x))
|
||||||
p2 = K.eval(K.foldr(lambda a, b: a*b, x))
|
p2 = K.eval(K.foldr(lambda a, b: a * b, x))
|
||||||
|
|
||||||
assert p1 < p2
|
assert p1 < p2
|
||||||
assert 9e-38 < p2 <= 1e-37
|
assert 9e-38 < p2 <= 1e-37
|
||||||
|
@ -9,6 +9,7 @@ from keras import backend as K
|
|||||||
from keras.models import model_from_json, model_from_yaml
|
from keras.models import model_from_json, model_from_yaml
|
||||||
from keras.utils.test_utils import keras_test
|
from keras.utils.test_utils import keras_test
|
||||||
|
|
||||||
|
|
||||||
@keras_test
|
@keras_test
|
||||||
def test_get_updates_for():
|
def test_get_updates_for():
|
||||||
a = Input(shape=(2,))
|
a = Input(shape=(2,))
|
||||||
@ -26,7 +27,7 @@ def test_get_losses_for():
|
|||||||
dense_layer = Dense(1)
|
dense_layer = Dense(1)
|
||||||
dense_layer.add_loss(0, inputs=a)
|
dense_layer.add_loss(0, inputs=a)
|
||||||
dense_layer.add_loss(1, inputs=None)
|
dense_layer.add_loss(1, inputs=None)
|
||||||
|
|
||||||
assert dense_layer.get_losses_for(a) == [0]
|
assert dense_layer.get_losses_for(a) == [0]
|
||||||
assert dense_layer.get_losses_for(None) == [1]
|
assert dense_layer.get_losses_for(None) == [1]
|
||||||
|
|
||||||
|
@ -15,7 +15,7 @@ class TestImage:
|
|||||||
gray_images = []
|
gray_images = []
|
||||||
for n in range(8):
|
for n in range(8):
|
||||||
bias = np.random.rand(img_w, img_h, 1) * 64
|
bias = np.random.rand(img_w, img_h, 1) * 64
|
||||||
variance = np.random.rand(img_w, img_h, 1) * (255-64)
|
variance = np.random.rand(img_w, img_h, 1) * (255 - 64)
|
||||||
imarray = np.random.rand(img_w, img_h, 3) * variance + bias
|
imarray = np.random.rand(img_w, img_h, 3) * variance + bias
|
||||||
im = Image.fromarray(imarray.astype('uint8')).convert('RGB')
|
im = Image.fromarray(imarray.astype('uint8')).convert('RGB')
|
||||||
rgb_images.append(im)
|
rgb_images.append(im)
|
||||||
|
@ -96,6 +96,7 @@ def test_ModelCheckpoint():
|
|||||||
os.remove(filepath.format(epoch=1))
|
os.remove(filepath.format(epoch=1))
|
||||||
os.remove(filepath.format(epoch=3))
|
os.remove(filepath.format(epoch=3))
|
||||||
|
|
||||||
|
|
||||||
def test_EarlyStopping():
|
def test_EarlyStopping():
|
||||||
(X_train, y_train), (X_test, y_test) = get_test_data(nb_train=train_samples,
|
(X_train, y_train), (X_test, y_test) = get_test_data(nb_train=train_samples,
|
||||||
nb_test=test_samples,
|
nb_test=test_samples,
|
||||||
|
@ -86,10 +86,10 @@ def test_identity(tensor_shape):
|
|||||||
if len(tensor_shape) > 2:
|
if len(tensor_shape) > 2:
|
||||||
with pytest.raises(Exception):
|
with pytest.raises(Exception):
|
||||||
_runner(initializations.identity, tensor_shape,
|
_runner(initializations.identity, tensor_shape,
|
||||||
target_mean=1./SHAPE[0], target_max=1.)
|
target_mean=1. / SHAPE[0], target_max=1.)
|
||||||
else:
|
else:
|
||||||
_runner(initializations.identity, tensor_shape,
|
_runner(initializations.identity, tensor_shape,
|
||||||
target_mean=1./SHAPE[0], target_max=1.)
|
target_mean=1. / SHAPE[0], target_max=1.)
|
||||||
|
|
||||||
|
|
||||||
@pytest.mark.parametrize('tensor_shape', [FC_SHAPE, CONV_SHAPE], ids=['FC', 'CONV'])
|
@pytest.mark.parametrize('tensor_shape', [FC_SHAPE, CONV_SHAPE], ids=['FC', 'CONV'])
|
||||||
|
@ -105,13 +105,13 @@ def test_top_k_categorical_accuracy():
|
|||||||
y_pred = K.variable(np.array([[0.3, 0.2, 0.1], [0.1, 0.2, 0.7]]))
|
y_pred = K.variable(np.array([[0.3, 0.2, 0.1], [0.1, 0.2, 0.7]]))
|
||||||
y_true = K.variable(np.array([[0, 1, 0], [1, 0, 0]]))
|
y_true = K.variable(np.array([[0, 1, 0], [1, 0, 0]]))
|
||||||
success_result = K.eval(metrics.top_k_categorical_accuracy(y_true, y_pred,
|
success_result = K.eval(metrics.top_k_categorical_accuracy(y_true, y_pred,
|
||||||
k=3))
|
k=3))
|
||||||
assert success_result == 1
|
assert success_result == 1
|
||||||
partial_result = K.eval(metrics.top_k_categorical_accuracy(y_true, y_pred,
|
partial_result = K.eval(metrics.top_k_categorical_accuracy(y_true, y_pred,
|
||||||
k=2))
|
k=2))
|
||||||
assert partial_result == 0.5
|
assert partial_result == 0.5
|
||||||
failure_result = K.eval(metrics.top_k_categorical_accuracy(y_true, y_pred,
|
failure_result = K.eval(metrics.top_k_categorical_accuracy(y_true, y_pred,
|
||||||
k=1))
|
k=1))
|
||||||
assert failure_result == 0
|
assert failure_result == 0
|
||||||
|
|
||||||
|
|
||||||
|
@ -49,6 +49,7 @@ def test_clasify_build_fn():
|
|||||||
|
|
||||||
def test_clasify_class_build_fn():
|
def test_clasify_class_build_fn():
|
||||||
class ClassBuildFnClf(object):
|
class ClassBuildFnClf(object):
|
||||||
|
|
||||||
def __call__(self, hidden_dims):
|
def __call__(self, hidden_dims):
|
||||||
return build_fn_clf(hidden_dims)
|
return build_fn_clf(hidden_dims)
|
||||||
|
|
||||||
@ -61,6 +62,7 @@ def test_clasify_class_build_fn():
|
|||||||
|
|
||||||
def test_clasify_inherit_class_build_fn():
|
def test_clasify_inherit_class_build_fn():
|
||||||
class InheritClassBuildFnClf(KerasClassifier):
|
class InheritClassBuildFnClf(KerasClassifier):
|
||||||
|
|
||||||
def __call__(self, hidden_dims):
|
def __call__(self, hidden_dims):
|
||||||
return build_fn_clf(hidden_dims)
|
return build_fn_clf(hidden_dims)
|
||||||
|
|
||||||
@ -110,6 +112,7 @@ def test_regression_build_fn():
|
|||||||
|
|
||||||
def test_regression_class_build_fn():
|
def test_regression_class_build_fn():
|
||||||
class ClassBuildFnReg(object):
|
class ClassBuildFnReg(object):
|
||||||
|
|
||||||
def __call__(self, hidden_dims):
|
def __call__(self, hidden_dims):
|
||||||
return build_fn_reg(hidden_dims)
|
return build_fn_reg(hidden_dims)
|
||||||
|
|
||||||
@ -122,6 +125,7 @@ def test_regression_class_build_fn():
|
|||||||
|
|
||||||
def test_regression_inherit_class_build_fn():
|
def test_regression_inherit_class_build_fn():
|
||||||
class InheritClassBuildFnReg(KerasRegressor):
|
class InheritClassBuildFnReg(KerasRegressor):
|
||||||
|
|
||||||
def __call__(self, hidden_dims):
|
def __call__(self, hidden_dims):
|
||||||
return build_fn_reg(hidden_dims)
|
return build_fn_reg(hidden_dims)
|
||||||
|
|
||||||
|
Loading…
Reference in New Issue
Block a user