Skip to content

Commit

Permalink
Enabled pep8 line lenght on the tensorflow backend. (#11514)
Browse files Browse the repository at this point in the history
  • Loading branch information
@gabrieldemarmiesse @fchollet
gabrieldemarmiesse authored and fchollet committed Oct 29, 2018
1 parent a37b41d commit 48c1c96
Show file tree
Hide file tree
Showing 2 changed files with 83 additions and 44 deletions.
124 changes: 82 additions & 42 deletions keras/backend/tensorflow_backend.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -255,7 +255,8 @@ def _is_current_explicit_device(device_type):
device_type: A string containing `GPU` or `CPU` (case-insensitive).
# Returns
A boolean indicating if the current device scope is explicitly set on the device type.
A boolean indicating if the current device
scope is explicitly set on the device type.
# Raises
ValueError: If the `device_type` string indicates an unsupported device.
Expand All @@ -282,8 +283,10 @@ def _get_available_gpus():
def _has_nchw_support():
"""Check whether the current scope supports NCHW ops.
TensorFlow does not support NCHW on CPU. Therefore we check if we are not explicitly put on
CPU, and have GPUs available. In this case there will be soft-placing on the GPU device.
TensorFlow does not support NCHW on CPU.
Therefore we check if we are not explicitly put on
CPU, and have GPUs available.
In this case there will be soft-placing on the GPU device.
# Returns
bool: if the current scope device placement would support nchw
Expand Down Expand Up @@ -453,19 +456,23 @@ def is_keras_tensor(x):
>>> K.is_keras_tensor(np_var) # A numpy array is not a symbolic tensor.
ValueError
>>> k_var = tf.placeholder('float32', shape=(1,1))
>>> K.is_keras_tensor(k_var) # A variable indirectly created outside of keras is not a Keras tensor.
>>> # A variable indirectly created outside of keras is not a Keras tensor.
>>> K.is_keras_tensor(k_var)
False
>>> keras_var = K.variable(np_var)
>>> K.is_keras_tensor(keras_var) # A variable created with the keras backend is not a Keras tensor.
>>> # A variable created with the keras backend is not a Keras tensor.
>>> K.is_keras_tensor(keras_var)
False
>>> keras_placeholder = K.placeholder(shape=(2, 4, 5))
>>> K.is_keras_tensor(keras_placeholder) # A placeholder is not a Keras tensor.
>>> # A placeholder is not a Keras tensor.
>>> K.is_keras_tensor(keras_placeholder)
False
>>> keras_input = Input([10])
>>> K.is_keras_tensor(keras_input) # An Input is a Keras tensor.
True
>>> keras_layer_output = Dense(10)(keras_input)
>>> K.is_keras_tensor(keras_layer_output) # Any Keras layer output is a Keras tensor.
>>> # Any Keras layer output is a Keras tensor.
>>> K.is_keras_tensor(keras_layer_output)
True
```
"""
Expand Down Expand Up @@ -1963,9 +1970,12 @@ def batch_normalization(x, mean, var, beta, gamma, axis=-1, epsilon=1e-3):
else:
tf_data_format = None

if tf_data_format == 'NHWC' or tf_data_format == 'NCHW' and _has_nchw_support():
if (tf_data_format == 'NHWC'
or tf_data_format == 'NCHW'
and _has_nchw_support()):
# The mean / var / beta / gamma may be processed by broadcast
# so it may have extra axes with 1, it is not needed and should be removed
# so it may have extra axes with 1,
# it is not needed and should be removed
if ndim(mean) > 1:
mean = tf.reshape(mean, [-1])
if ndim(var) > 1:
Expand Down Expand Up @@ -2063,7 +2073,8 @@ def resize_images(x,
A tensor.
# Raises
ValueError: if `data_format` is neither `"channels_last"` or `"channels_first"`.
ValueError: if `data_format` is
neither `"channels_last"` or `"channels_first"`.
"""
if data_format == 'channels_first':
rows, cols = 2, 3
Expand Down Expand Up @@ -2115,7 +2126,8 @@ def resize_volumes(x, depth_factor, height_factor, width_factor, data_format):
A tensor.
# Raises
ValueError: if `data_format` is neither `"channels_last"` or `"channels_first"`.
ValueError: if `data_format` is
neither `"channels_last"` or `"channels_first"`.
"""
if data_format == 'channels_first':
output = repeat_elements(x, depth_factor, axis=2)
Expand Down Expand Up @@ -2334,7 +2346,8 @@ def spatial_2d_padding(x, padding=((1, 1), (1, 1)), data_format=None):
A padded 4D tensor.
# Raises
ValueError: if `data_format` is neither `"channels_last"` or `"channels_first"`.
ValueError: if `data_format` is
neither `"channels_last"` or `"channels_first"`.
"""
assert len(padding) == 2
assert len(padding[0]) == 2
Expand Down Expand Up @@ -2369,7 +2382,8 @@ def spatial_3d_padding(x, padding=((1, 1), (1, 1), (1, 1)), data_format=None):
A padded 5D tensor.
# Raises
ValueError: if `data_format` is neither `"channels_last"` or `"channels_first"`.
ValueError: if `data_format` is
neither `"channels_last"` or `"channels_first"`.
"""
assert len(padding) == 3
Expand Down Expand Up @@ -2447,9 +2461,10 @@ def slice(x, start, size):
along each axis.
# Returns
Tensor `x[start[0]: start[0] + size[0],
...,
start[-1]: start[-1] + size[-1]]`
A sliced tensor:
```python
new_x = x[start[0]: start[0] + size[0], ..., start[-1]: start[-1] + size[-1]]
```
"""
return tf.slice(x, start, size)

Expand Down Expand Up @@ -2766,13 +2781,15 @@ def __call__(self, inputs):

# callable generated by Session._make_callable_from_options accepts
# `run_metadata` keyword argument since TF 1.10
if (self.run_metadata and
StrictVersion(tf.__version__.split('-')[0]) < StrictVersion('1.10.0')):
if py_any(is_tensor(x) for x in inputs):
raise ValueError(
'In order to feed symbolic tensors to a Keras model and set '
'`run_metadata`, you need tensorflow 1.10 or higher.')
return self._legacy_call(inputs)
if self.run_metadata:
current_version = StrictVersion(tf.__version__.split('-')[0])
if current_version < StrictVersion('1.10.0'):
if py_any(is_tensor(x) for x in inputs):
raise ValueError(
'In order to feed symbolic tensors '
'to a Keras model and set '
'`run_metadata`, you need tensorflow 1.10 or higher.')
return self._legacy_call(inputs)

return self._call(inputs)
else:
Expand Down Expand Up @@ -2800,9 +2817,11 @@ def function(inputs, outputs, updates=None, **kwargs):
"""
if kwargs:
for key in kwargs:
if not (has_arg(tf.Session.run, key, True) or has_arg(Function.__init__, key, True)):
msg = 'Invalid argument "%s" passed to K.function with TensorFlow backend' % key
raise ValueError(msg)
session_has_key = has_arg(tf.Session.run, key, True)
function_has_key = has_arg(Function.__init__, key, True)
if not (session_has_key or function_has_key):
raise ValueError('Invalid argument "%s" passed to K.function '
'with TensorFlow backend' % key)
return Function(inputs, outputs, updates=updates, **kwargs)


Expand Down Expand Up @@ -3034,10 +3053,12 @@ def _step(time, output_ta_t, *states):
tiled_mask_t = tf.tile(mask_t,
tf.stack([1, tf.shape(output)[1]]))
output = tf.where(tiled_mask_t, output, states[0])
new_states = [
tf.where(tf.tile(mask_t, tf.stack([1, tf.shape(new_states[i])[1]])),
new_states[i], states[i]) for i in range(len(states))
]
tmp = []
for i in range(len(states)):
multiples = tf.stack([1, tf.shape(new_states[i])[1]])
tiled = tf.tile(mask_t, multiples)
tmp.append(tf.where(tiled, new_states[i], states[i]))
new_states = tmp
output_ta_t = output_ta_t.write(time, output)
return (time + 1, output_ta_t) + tuple(new_states)
else:
Expand Down Expand Up @@ -3139,7 +3160,8 @@ def else_expression_fn():
condition = tf.reshape(condition, cond_shape)
expr_shape = tf.shape(then_expression)
shape_diff = expr_shape - cond_shape
tile_shape = tf.where(shape_diff > 0, expr_shape, tf.ones_like(expr_shape))
zero_expr_shape = tf.ones_like(expr_shape)
tile_shape = tf.where(shape_diff > 0, expr_shape, zero_expr_shape)
condition = tf.tile(condition, tile_shape)
x = tf.where(condition, then_expression, else_expression)
return x
Expand Down Expand Up @@ -4020,7 +4042,8 @@ def pool2d(x, pool_size, strides=(1, 1),
A tensor, result of 2D pooling.
# Raises
ValueError: if `data_format` is neither `"channels_last"` or `"channels_first"`.
ValueError: if `data_format` is
neither `"channels_last"` or `"channels_first"`.
ValueError: if `pool_mode` is neither `"max"` or `"avg"`.
"""
data_format = normalize_data_format(data_format)
Expand Down Expand Up @@ -4066,7 +4089,8 @@ def pool3d(x, pool_size, strides=(1, 1, 1), padding='valid',
A tensor, result of 3D pooling.
# Raises
ValueError: if `data_format` is neither `"channels_last"` or `"channels_first"`.
ValueError: if `data_format` is
neither `"channels_last"` or `"channels_first"`.
ValueError: if `pool_mode` is neither `"max"` or `"avg"`.
"""
data_format = normalize_data_format(data_format)
Expand Down Expand Up @@ -4117,7 +4141,8 @@ def bias_add(x, bias, data_format=None):
data_format = normalize_data_format(data_format)
bias_shape = int_shape(bias)
if len(bias_shape) != 1 and len(bias_shape) != ndim(x) - 1:
raise ValueError('Unexpected bias dimensions %d, expect to be 1 or %d dimensions'
raise ValueError('Unexpected bias dimensions %d, '
'expect to be 1 or %d dimensions'
% (len(bias_shape), ndim(x)))
if ndim(x) == 5:
if len(bias_shape) == 1:
Expand Down Expand Up @@ -4281,14 +4306,18 @@ def range_less_than(_, current_input):
label_shape)
label_ind = tf.boolean_mask(label_array, dense_mask)

batch_array = tf.transpose(tf.reshape(tf.tile(tf.range(label_shape[0]),
max_num_labels_tns), reverse(label_shape, 0)))
tmp = tf.tile(tf.range(label_shape[0]), max_num_labels_tns)
batch_array = tf.transpose(tf.reshape(tmp, reverse(label_shape, 0)))
batch_ind = tf.boolean_mask(batch_array, dense_mask)
indices = tf.transpose(tf.reshape(concatenate([batch_ind, label_ind], axis=0), [2, -1]))

indices = concatenate([batch_ind, label_ind], axis=0)
indices = tf.transpose(tf.reshape(indices, [2, -1]))

vals_sparse = tf.gather_nd(labels, indices)

return tf.SparseTensor(tf.to_int64(indices), vals_sparse, tf.to_int64(label_shape))
indices = tf.to_int64(indices)
label_shape = tf.to_int64(label_shape)
return tf.SparseTensor(indices, vals_sparse, label_shape)


def ctc_batch_cost(y_true, y_pred, input_length, label_length):
Expand Down Expand Up @@ -4361,8 +4390,13 @@ def ctc_decode(y_pred, input_length, greedy=True, beam_width=100,
sequence_length=input_length, beam_width=beam_width,
top_paths=top_paths)

decoded_dense = [tf.sparse_to_dense(st.indices, st.dense_shape, st.values, default_value=-1)
for st in decoded]
decoded_dense = []
for st in decoded:
dense_tensor = tf.sparse_to_dense(st.indices,
st.dense_shape,
st.values,
default_value=-1)
decoded_dense.append(dense_tensor)
return (decoded_dense, log_prob)


Expand Down Expand Up @@ -4429,7 +4463,8 @@ def local_conv1d(inputs, kernel, kernel_size, strides, data_format=None):
data_format: the data format, channels_first or channels_last
# Returns
the tensor after 1d conv with un-shared weights, with shape (batch_size, output_length, filters)
the tensor after 1d conv with un-shared weights,
with shape (batch_size, output_length, filters)
# Raises
ValueError: If `data_format` is neither
Expand All @@ -4453,7 +4488,12 @@ def local_conv1d(inputs, kernel, kernel_size, strides, data_format=None):
return permute_dimensions(output, (1, 0, 2))


def local_conv2d(inputs, kernel, kernel_size, strides, output_shape, data_format=None):
def local_conv2d(inputs,
kernel,
kernel_size,
strides,
output_shape,
data_format=None):
"""Apply 2D conv with un-shared weights.
# Arguments
Expand Down
3 changes: 1 addition & 2 deletions pytest.ini
View file
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,6 @@ timeout = 720

pep8ignore=* E402 \
* E731 \
* W503 \
keras/backend/tensorflow_backend.py E501
* W503
# Enable line length testing with maximum line length of 85
pep8maxlinelength = 85

0 comments on commit 48c1c96

Please sign in to comment.