def uniform(name, shape, scale=0.05, dtype=tf.sg_floatx, summary=True):
r"""Creates a tensor variable of which initial values are
random numbers based on uniform distribution.
Note that the default value of `scale` (=0.05) is different from
the min/max values (=0.0, 1.0) of tf.random_uniform_initializer.
Args:
name: The name of the new variable.
shape: A tuple/list of integers or an integer.
If shape is an integer, it's converted to a list.
scale: A Python scalar. All initial values should be in range `[-scale, scale)`. Default is .05.
dtype: The data type. Only floating point types are supported. Default is float32.
summary: If True, add this constant to tensor board summary.
Returns:
A `Variable`.
"""
shape = shape if isinstance(shape, (tuple, list)) else [shape]
x = tf.get_variable(name,
shape,
dtype=dtype,
initializer=tf.random_uniform_initializer(
minval=-scale, maxval=scale))
# add summary
if not tf.get_variable_scope().reuse and summary:
tf.sg_summary_param(x)
return x
def orthogonal(name, shape, scale=1.1, dtype=tf.sg_floatx, summary=True):
r"""Creates a tensor variable of which initial values are of
an orthogonal ndarray.
See [Saxe et al. 2014.](http://arxiv.org/pdf/1312.6120.pdf)
Args:
name: The name of new variable.
shape: A tuple/list of integers.
scale: A Python scalar.
dtype: Either float32 or float64.
summary: If True, add this constant to tensor board summary.
Returns:
A `Variable`.
"""
flat_shape = (shape[0], np.prod(shape[1:]))
a = np.random.normal(0.0, 1.0, flat_shape)
u, _, v = np.linalg.svd(a, full_matrices=False)
# pick the one with the correct shape
q = u if u.shape == flat_shape else v
q = q.reshape(shape)
# create variable
x = tf.get_variable(name,
initializer=tf.constant(scale *
q[:shape[0], :shape[1]],
dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse and summary:
tf.sg_summary_param(x)
return x
def external(name, value, dtype=tf.sg_floatx, summary=True):
r"""Creates a tensor variable of which initial values are `value`.
For example,
```
external("external", [3,3,1,2])
=> [3. 3. 1. 2.]
```
Args:
name: The name of new variable.
value: A constant value (or list) of output type `dtype`.
dtype: The type of the elements of the resulting tensor.
summary: If True, add this constant to tensor board summary.
Returns:
A `Variable`. Has the same contents as `value` of `dtype`.
"""
# create variable
x = tf.get_variable(name, initializer=tf.constant(value, dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse and summary:
tf.sg_summary_param(x)
return x
def constant(name, shape, value=0, dtype=tf.sg_floatx, summary=True, regularizer=None, trainable=True):
r"""Creates a tensor variable of which initial values are `value` and shape is `shape`.
Args:
name: The name of new variable.
shape: A tuple/list of integers or an integer.
If shape is an integer, it is converted to a list.
value: A Python scalar. All elements of the initialized variable
will be set to this value. Default is 0.
dtype: The data type. Only floating point types are supported. Default is float32.
summary: If True, add this constant to tensor board summary.
regularizer: A (Tensor -> Tensor or None) function; the result of applying it on a newly created variable
will be added to the collection tf.GraphKeys.REGULARIZATION_LOSSES and can be used for regularization
trainable: If True, add this constant to trainable collection. Default is True.
Returns:
A `Variable`.
"""
shape = shape if isinstance(shape, (tuple, list)) else [shape]
x = tf.get_variable(name, shape, dtype=dtype,
initializer=tf.constant_initializer(value),
regularizer=regularizer, trainable=trainable)
# add summary
if summary:
tf.sg_summary_param(x)
return x
def identity(name, dim, scale=1, dtype=tf.sg_floatx, summary=True):
r"""Creates a tensor variable of which initial values are of
an identity matrix.
Note that the default value of `scale` (=0.05) is different from
the min/max values (=0.0, 1.0) of tf.random_uniform_initializer.
For example,
```
identity("identity", 3, 2) =>
[[2. 0. 0.]
[0. 2. 0.]
[0. 0. 2.]]
```
Args:
name: The name of new variable.
dim: An int. The size of the first and second dimension of the output tensor.
scale: A Python scalar. The value on the diagonal.
dtype: The type of the elements of the resulting tensor.
summary: If True, add this constant to tensor board summary.
Returns:
A 2-D `Variable`.
"""
x = tf.get_variable(name,
initializer=tf.constant(np.eye(dim) * scale,
dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse and summary:
tf.sg_summary_param(x)
return x
def external(name, value, dtype=tf.sg_floatx, summary=True, regularizer=None, trainable=True):
r"""Creates a tensor variable of which initial values are `value`.
For example,
```
external("external", [3,3,1,2])
=> [3. 3. 1. 2.]
```
Args:
name: The name of new variable.
value: A constant value (or list) of output type `dtype`.
dtype: The type of the elements of the resulting tensor.
summary: If True, add this constant to tensor board summary.
regularizer: A (Tensor -> Tensor or None) function; the result of applying it on a newly created variable
will be added to the collection tf.GraphKeys.REGULARIZATION_LOSSES and can be used for regularization
trainable: If True, add this constant to trainable collection. Default is True.
Returns:
A `Variable`. Has the same contents as `value` of `dtype`.
"""
# create variable
x = tf.get_variable(name,
initializer=tf.constant(value, dtype=dtype),
regularizer=regularizer, trainable=trainable)
# add summary
if summary:
tf.sg_summary_param(x)
return x
def uniform(name, shape, scale=0.05, dtype=tf.sg_floatx, summary=True, regularizer=None, trainable=True):
r"""Creates a tensor variable of which initial values are
random numbers based on uniform distribution.
Note that the default value of `scale` (=0.05) is different from
the min/max values (=0.0, 1.0) of tf.random_uniform_initializer.
Args:
name: The name of the new variable.
shape: A tuple/list of integers or an integer.
If shape is an integer, it's converted to a list.
scale: A Python scalar. All initial values should be in range `[-scale, scale)`. Default is .05.
dtype: The data type. Only floating point types are supported. Default is float32.
summary: If True, add this constant to tensor board summary.
regularizer: A (Tensor -> Tensor or None) function; the result of applying it on a newly created variable
will be added to the collection tf.GraphKeys.REGULARIZATION_LOSSES and can be used for regularization
trainable: If True, add this constant to trainable collection. Default is True.
Returns:
A `Variable`.
"""
shape = shape if isinstance(shape, (tuple, list)) else [shape]
x = tf.get_variable(name, shape, dtype=dtype,
initializer=tf.random_uniform_initializer(minval=-scale, maxval=scale),
regularizer=regularizer, trainable=trainable)
# add summary
if summary:
tf.sg_summary_param(x)
return x
def orthogonal(name, shape, scale=1.1, dtype=tf.sg_floatx):
r"""Returns a random orthogonal initializer.
See Saxe et al. 2014 `http://arxiv.org/pdf/1312.6120.pdf`
Args:
name: A string. The name of the new or existing variable.
shape: A list or tuple of integers.
scale: A Python scalr.
dtype = A float32 or float64.
Returns:
A `Tensor` variable.
"""
flat_shape = (shape[0], np.prod(shape[1:]))
a = np.random.normal(0.0, 1.0, flat_shape)
u, _, v = np.linalg.svd(a, full_matrices=False)
# pick the one with the correct shape
q = u if u.shape == flat_shape else v
q = q.reshape(shape)
# create variable
x = tf.get_variable(name,
initializer=tf.constant(scale *
q[:shape[0], :shape[1]],
dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def orthogonal(name, shape, scale=1.1, dtype=tf.sg_floatx, summary=True, regularizer=None, trainable=True):
r"""Creates a tensor variable of which initial values are of
an orthogonal ndarray.
See [Saxe et al. 2014.](http://arxiv.org/pdf/1312.6120.pdf)
Args:
name: The name of new variable.
shape: A tuple/list of integers.
scale: A Python scalar.
dtype: Either float32 or float64.
summary: If True, add this constant to tensor board summary.
regularizer: A (Tensor -> Tensor or None) function; the result of applying it on a newly created variable
will be added to the collection tf.GraphKeys.REGULARIZATION_LOSSES and can be used for regularization
trainable: If True, add this constant to trainable collection. Default is True.
Returns:
A `Variable`.
"""
flat_shape = (shape[0], np.prod(shape[1:]))
a = np.random.normal(0.0, 1.0, flat_shape)
u, _, v = np.linalg.svd(a, full_matrices=False)
# pick the one with the correct shape
q = u if u.shape == flat_shape else v
q = q.reshape(shape)
# create variable
x = tf.get_variable(name,
initializer=tf.constant(scale * q[:shape[0], :shape[1]], dtype=dtype),
regularizer=regularizer, trainable=trainable)
# add summary
if summary:
tf.sg_summary_param(x)
return x
def identity(name, dim, scale=1, dtype=tf.sg_floatx):
x = tf.get_variable(name,
initializer=tf.constant(np.eye(dim) * scale, dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def constant(name, shape, value=0, dtype=tf.sg_floatx, summary=True):
r"""Creates a tensor variable of which initial values are `value` and shape is `shape`.
Args:
name: The name of new variable.
shape: A tuple/list of integers or an integer.
If shape is an integer, it is converted to a list.
value: A Python scalar. All elements of the initialized variable
will be set to this value. Default is 0.
dtype: The data type. Only floating point types are supported. Default is float32.
summary: If True, add this constant to tensor board summary.
Returns:
A `Variable`.
"""
shape = shape if isinstance(shape, (tuple, list)) else [shape]
x = tf.get_variable(name,
shape,
dtype=dtype,
initializer=tf.constant_initializer(value))
# add summary
if not tf.get_variable_scope().reuse and summary:
tf.sg_summary_param(x)
return x
def uniform(name, shape, scale=0.05, dtype=tf.sg_floatx):
shape = shape if isinstance(shape, (tuple, list)) else [shape]
x = tf.get_variable(name, shape, dtype=dtype,
initializer=tf.random_uniform_initializer(minval=-scale, maxval=scale))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def external(name, value, dtype=tf.sg_floatx):
# create variable
x = tf.get_variable(name,
initializer=tf.constant(value, dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def constant(name, shape, value=0, dtype=tf.sg_floatx):
shape = shape if isinstance(shape, (tuple, list)) else [shape]
x = tf.get_variable(name, shape, dtype=dtype,
initializer=tf.constant_initializer(value))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def constant(name, shape, value=0, dtype=tf.sg_floatx):
r"""Returns an initializer of `shape` with all elements set to a scalar `value`.
"""
shape = shape if isinstance(shape, (tuple, list)) else [shape]
x = tf.get_variable(name,
shape,
dtype=dtype,
initializer=tf.constant_initializer(value))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def init_custom_embeddings(name, embeddings_matrix, summary=True, trainable=False):
"""
Initializes the embedding vector with custom preloaded embeddings
"""
embedding = np.array(embeddings_matrix)
emb = tf.get_variable(name, shape=embedding.shape, initializer=tf.constant_initializer(embedding),
trainable=trainable)
if summary:
tf.sg_summary_param(emb)
return emb
def init_custom_embeddings(name, embeddings_matrix, summary=True, trainable=False):
"""
Initializes the embedding vector with custom preloaded embeddings
"""
embedding = np.array(embeddings_matrix)
emb = tf.get_variable(name, shape=embedding.shape, initializer=tf.constant_initializer(embedding),
trainable=trainable)
if summary:
tf.sg_summary_param(emb)
return emb
def uniform(name, shape, scale=0.05, dtype=tf.sg_floatx):
r"""Returns an initializer of random numbers based on uniform distribution.
Note that the default value of `scale` (=0.05) is different from
the min/max values (=0.0, 1.0) of tf.random_uniform_initializer.
"""
shape = shape if isinstance(shape, (tuple, list)) else [shape]
x = tf.get_variable(name,
shape,
dtype=dtype,
initializer=tf.random_uniform_initializer(
minval=-scale, maxval=scale))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def orthogonal(name, shape, scale=1.1, dtype=tf.sg_floatx):
# Sax et aE. ( http://arxiv.org/pdf/1312.6120.pdf )
flat_shape = (shape[0], np.prod(shape[1:]))
a = np.random.normal(0.0, 1.0, flat_shape)
u, _, v = np.linalg.svd(a, full_matrices=False)
# pick the one with the correct shape
q = u if u.shape == flat_shape else v
q = q.reshape(shape)
# create variable
x = tf.get_variable(name,
initializer=tf.constant(scale * q[:shape[0], :shape[1]], dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def external(name, value, dtype=tf.sg_floatx):
r"""Returns an initializer of `value`.
Args:
name: A string. The name of the new or existing variable.
value: A constant value (or array) of output type `dtype`.
dtype: The type of the elements of the resulting tensor. (optional)
Returns:
A `Tensor` variable.
"""
# create variable
x = tf.get_variable(name, initializer=tf.constant(value, dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
def identity(name,
dim,
scale=1,
dtype=tf.sg_floatx,
summary=True,
regularizer=None,
trainable=True):
r"""Creates a tensor variable of which initial values are of
an identity matrix.
Note that the default value of `scale` (=0.05) is different from
the min/max values (=0.0, 1.0) of tf.random_uniform_initializer.
For example,
```
identity("identity", 3, 2) =>
[[2. 0. 0.]
[0. 2. 0.]
[0. 0. 2.]]
```
Args:
name: The name of new variable.
dim: An int. The size of the first and second dimension of the output tensor.
scale: A Python scalar. The value on the diagonal.
dtype: The type of the elements of the resulting tensor.
summary: If True, add this constant to tensor board summary.
regularizer: A (Tensor -> Tensor or None) function; the result of applying it on a newly created variable
will be added to the collection tf.GraphKeys.REGULARIZATION_LOSSES and can be used for regularization
trainable: If True, add this constant to trainable collection. Default is True.
Returns:
A 2-D `Variable`.
"""
x = tf.get_variable(name,
initializer=tf.constant(np.eye(dim) * scale,
dtype=dtype),
regularizer=regularizer,
trainable=trainable)
# add summary
if summary:
tf.sg_summary_param(x)
return x
def identity(name, dim, scale=1, dtype=tf.sg_floatx):
r"""Returns an initializer of a 2-D identity tensor.
Args:
name: A string. The name of the new or existing variable.
dim: An int. The size of the first and second dimension of the output tensor
scale: An int (optional). The value on the diagonal.
shape: Shape of the new or existing variable.
dtype: A tensor datatype.
Returns:
A 2-D tensor variable with the value of `scale` on the diagoanl and zeros elsewhere.
"""
x = tf.get_variable(name,
initializer=tf.constant(np.eye(dim) * scale,
dtype=dtype))
# add summary
if not tf.get_variable_scope().reuse:
tf.sg_summary_param(x)
return x
