tf.Variable

View source on GitHub

See the variable guide.

tf.Variable(
    initial_value=None,
    trainable=None,
    validate_shape=True,
    caching_device=None,
    name=None,
    variable_def=None,
    dtype=None,
    import_scope=None,
    constraint=None,
    synchronization=tf.VariableSynchronization.AUTO,
    aggregation=tf.compat.v1.VariableAggregation.NONE,
    shape=None
)

A variable maintains shared, persistent state manipulated by a program.

The Variable() constructor requires an initial value for the variable, which can be a Tensor of any type and shape. This initial value defines the type and shape of the variable. After construction, the type and shape of the variable are fixed. The value can be changed using one of the assign methods.

v = tf.Variable(1.)
v.assign(2.)
<tf.Variable ... shape=() dtype=float32, numpy=2.0>
v.assign_add(0.5)
<tf.Variable ... shape=() dtype=float32, numpy=2.5>

The shape argument to Variable's constructor allows you to construct a variable with a less defined shape than its initial_value:

v = tf.Variable(1., shape=tf.TensorShape(None))
v.assign([[1.]])
<tf.Variable ... shape=<unknown> dtype=float32, numpy=array([[1.]], ...)>

Just like any Tensor, variables created with Variable() can be used as inputs to operations. Additionally, all the operators overloaded for the Tensor class are carried over to variables.

w = tf.Variable([[1.], [2.]])
x = tf.constant([[3., 4.]])
tf.matmul(w, x)
<tf.Tensor:... shape=(2, 2), ... numpy=
  array([[3., 4.],
         [6., 8.]], dtype=float32)>
tf.sigmoid(w + x)
<tf.Tensor:... shape=(2, 2), ...>

When building a machine learning model it is often convenient to distinguish between variables holding trainable model parameters and other variables such as a step variable used to count training steps. To make this easier, the variable constructor supports a trainable=<bool> parameter. tf.GradientTape watches trainable variables by default:

with tf.GradientTape(persistent=True) as tape:
  trainable = tf.Variable(1.)
  non_trainable = tf.Variable(2., trainable=False)
  x1 = trainable * 2.
  x2 = non_trainable * 3.
tape.gradient(x1, trainable)
<tf.Tensor:... shape=(), dtype=float32, numpy=2.0>
assert tape.gradient(x2, non_trainable) is None  # Unwatched

Variables are automatically tracked when assigned to attributes of types inheriting from tf.Module.

m = tf.Module()
m.v = tf.Variable([1.])
m.trainable_variables
(<tf.Variable ... shape=(1,) ... numpy=array([1.], dtype=float32)>,)

This tracking then allows saving variable values to training checkpoints, or to SavedModels which include serialized TensorFlow graphs.

Variables are often captured and manipulated by tf.functions. This works the same way the un-decorated function would have:

v = tf.Variable(0.)
read_and_decrement = tf.function(lambda: v.assign_sub(0.1))
read_and_decrement()
<tf.Tensor: shape=(), dtype=float32, numpy=-0.1>
read_and_decrement()
<tf.Tensor: shape=(), dtype=float32, numpy=-0.2>

Variables created inside a tf.function must be owned outside the function and be created only once:

class M(tf.Module):
  @tf.function
  def __call__(self, x):
    if not hasattr(self, "v"):  # Or set self.v to None in __init__
      self.v = tf.Variable(x)
    return self.v * x
m = M()
m(2.)
<tf.Tensor: shape=(), dtype=float32, numpy=4.0>
m(3.)
<tf.Tensor: shape=(), dtype=float32, numpy=6.0>
m.v
<tf.Variable ... shape=() dtype=float32, numpy=2.0>

See the tf.function documentation for details.

Child Classes

class SaveSliceInfo

Methods

assign

View source

assign(
    value, use_locking=False, name=None, read_value=True
)

Assigns a new value to the variable.

This is essentially a shortcut for assign(self, value).

assign_add

View source

assign_add(
    delta, use_locking=False, name=None, read_value=True
)

Adds a value to this variable.

This is essentially a shortcut for assign_add(self, delta).

assign_sub

View source

assign_sub(
    delta, use_locking=False, name=None, read_value=True
)

Subtracts a value from this variable.

This is essentially a shortcut for assign_sub(self, delta).

batch_scatter_update

View source

batch_scatter_update(
    sparse_delta, use_locking=False, name=None
)

Assigns tf.IndexedSlices to this variable batch-wise.

Analogous to batch_gather. This assumes that this variable and the sparse_delta IndexedSlices have a series of leading dimensions that are the same for all of them, and the updates are performed on the last dimension of indices. In other words, the dimensions should be the following:

num_prefix_dims = sparse_delta.indices.ndims - 1 batch_dim = num_prefix_dims + 1 sparse_delta.updates.shape = sparse_delta.indices.shape + var.shape[ batch_dim:]

where

sparse_delta.updates.shape[:num_prefix_dims] == sparse_delta.indices.shape[:num_prefix_dims] == var.shape[:num_prefix_dims]

And the operation performed can be expressed as:

var[i_1, ..., i_n, sparse_delta.indices[i_1, ..., i_n, j]] = sparse_delta.updates[ i_1, ..., i_n, j]

When sparse_delta.indices is a 1D tensor, this operation is equivalent to scatter_update.

To avoid this operation one can looping over the first ndims of the variable and using scatter_update on the subtensors that result of slicing the first dimension. This is a valid option for ndims = 1, but less efficient than this implementation.

count_up_to

View source

count_up_to(
    limit
)

Increments this variable until it reaches limit. (deprecated)

When that Op is run it tries to increment the variable by 1. If incrementing the variable would bring it above limit then the Op raises the exception OutOfRangeError.

If no error is raised, the Op outputs the value of the variable before the increment.

This is essentially a shortcut for count_up_to(self, limit).

eval

View source

eval(
    session=None
)

In a session, computes and returns the value of this variable.

This is not a graph construction method, it does not add ops to the graph.

This convenience method requires a session where the graph containing this variable has been launched. If no session is passed, the default session is used. See tf.compat.v1.Session for more information on launching a graph and on sessions.

v = tf.Variable([1, 2])
init = tf.compat.v1.global_variables_initializer()

with tf.compat.v1.Session() as sess:
    sess.run(init)
    # Usage passing the session explicitly.
    print(v.eval(sess))
    # Usage with the default session.  The 'with' block
    # above makes 'sess' the default session.
    print(v.eval())

experimental_ref

View source

experimental_ref()

DEPRECATED FUNCTION

from_proto

View source

@staticmethod
from_proto(
    variable_def, import_scope=None
)

Returns a Variable object created from variable_def.

gather_nd

View source

gather_nd(
    indices, name=None
)

Gather slices from params into a Tensor with shape specified by indices.

See tf.gather_nd for details.

get_shape

View source

get_shape()

Alias of Variable.shape.

initialized_value

View source

initialized_value()

Returns the value of the initialized variable. (deprecated)

You should use this instead of the variable itself to initialize another variable with a value that depends on the value of this variable.

# Initialize 'v' with a random tensor.
v = tf.Variable(tf.random.truncated_normal([10, 40]))
# Use `initialized_value` to guarantee that `v` has been
# initialized before its value is used to initialize `w`.
# The random values are picked only once.
w = tf.Variable(v.initialized_value() * 2.0)

load

View source

load(
    value, session=None
)

Load new value into this variable. (deprecated)

Writes new value to variable's memory. Doesn't add ops to the graph.

This convenience method requires a session where the graph containing this variable has been launched. If no session is passed, the default session is used. See tf.compat.v1.Session for more information on launching a graph and on sessions.

v = tf.Variable([1, 2])
init = tf.compat.v1.global_variables_initializer()

with tf.compat.v1.Session() as sess:
    sess.run(init)
    # Usage passing the session explicitly.
    v.load([2, 3], sess)
    print(v.eval(sess)) # prints [2 3]
    # Usage with the default session.  The 'with' block
    # above makes 'sess' the default session.
    v.load([3, 4], sess)
    print(v.eval()) # prints [3 4]

read_value

View source

read_value()

Returns the value of this variable, read in the current context.

Can be different from value() if it's on another device, with control dependencies, etc.

ref

View source

ref()

Returns a hashable reference object to this Variable.

The primary use case for this API is to put variables in a set/dictionary. We can't put variables in a set/dictionary as variable.__hash__() is no longer available starting Tensorflow 2.0.

The following will raise an exception starting 2.0

x = tf.Variable(5)
y = tf.Variable(10)
z = tf.Variable(10)
variable_set = {x, y, z}
Traceback (most recent call last):

TypeError: Variable is unhashable. Instead, use tensor.ref() as the key.
variable_dict = {x: 'five', y: 'ten'}
Traceback (most recent call last):

TypeError: Variable is unhashable. Instead, use tensor.ref() as the key.

Instead, we can use variable.ref().

variable_set = {x.ref(), y.ref(), z.ref()}
x.ref() in variable_set
True
variable_dict = {x.ref(): 'five', y.ref(): 'ten', z.ref(): 'ten'}
variable_dict[y.ref()]
'ten'

Also, the reference object provides .deref() function that returns the original Variable.

x = tf.Variable(5)
x.ref().deref()
<tf.Variable 'Variable:0' shape=() dtype=int32, numpy=5>

scatter_add

View source

scatter_add(
    sparse_delta, use_locking=False, name=None
)

Adds tf.IndexedSlices to this variable.

scatter_div

View source

scatter_div(
    sparse_delta, use_locking=False, name=None
)

Divide this variable by tf.IndexedSlices.

scatter_max

View source

scatter_max(
    sparse_delta, use_locking=False, name=None
)

Updates this variable with the max of tf.IndexedSlices and itself.

scatter_min

View source

scatter_min(
    sparse_delta, use_locking=False, name=None
)

Updates this variable with the min of tf.IndexedSlices and itself.

scatter_mul

View source

scatter_mul(
    sparse_delta, use_locking=False, name=None
)

Multiply this variable by tf.IndexedSlices.

scatter_nd_add

View source

scatter_nd_add(
    indices, updates, name=None
)

Applies sparse addition to individual values or slices in a Variable.

The Variable has rank P and indices is a Tensor of rank Q.

indices must be integer tensor, containing indices into self. It must be shape [d_0, ..., d_{Q-2}, K] where 0 < K <= P.

The innermost dimension of indices (with length K) corresponds to indices into elements (if K = P) or slices (if K < P) along the Kth dimension of self.

updates is Tensor of rank Q-1+P-K with shape:

[d_0, ..., d_{Q-2}, self.shape[K], ..., self.shape[P-1]].

For example, say we want to add 4 scattered elements to a rank-1 tensor to 8 elements. In Python, that update would look like this:

    v = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
    indices = tf.constant([[4], [3], [1] ,[7]])
    updates = tf.constant([9, 10, 11, 12])
    v.scatter_nd_add(indices, updates)
    print(v)

The resulting update to v would look like this:

[1, 13, 3, 14, 14, 6, 7, 20]

See tf.scatter_nd for more details about how to make updates to slices.

scatter_nd_sub

View source

scatter_nd_sub(
    indices, updates, name=None
)

Applies sparse subtraction to individual values or slices in a Variable.

Assuming the variable has rank P and indices is a Tensor of rank Q.

indices must be integer tensor, containing indices into self. It must be shape [d_0, ..., d_{Q-2}, K] where 0 < K <= P.

The innermost dimension of indices (with length K) corresponds to indices into elements (if K = P) or slices (if K < P) along the Kth dimension of self.

updates is Tensor of rank Q-1+P-K with shape:

[d_0, ..., d_{Q-2}, self.shape[K], ..., self.shape[P-1]].

For example, say we want to add 4 scattered elements to a rank-1 tensor to 8 elements. In Python, that update would look like this:

    v = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
    indices = tf.constant([[4], [3], [1] ,[7]])
    updates = tf.constant([9, 10, 11, 12])
    v.scatter_nd_sub(indices, updates)
    print(v)

After the update v would look like this:

[1, -9, 3, -6, -4, 6, 7, -4]

See tf.scatter_nd for more details about how to make updates to slices.

scatter_nd_update

View source

scatter_nd_update(
    indices, updates, name=None
)

Applies sparse assignment to individual values or slices in a Variable.

The Variable has rank P and indices is a Tensor of rank Q.

indices must be integer tensor, containing indices into self. It must be shape [d_0, ..., d_{Q-2}, K] where 0 < K <= P.

The innermost dimension of indices (with length K) corresponds to indices into elements (if K = P) or slices (if K < P) along the Kth dimension of self.

updates is Tensor of rank Q-1+P-K with shape:

[d_0, ..., d_{Q-2}, self.shape[K], ..., self.shape[P-1]].

For example, say we want to add 4 scattered elements to a rank-1 tensor to 8 elements. In Python, that update would look like this:

    v = tf.Variable([1, 2, 3, 4, 5, 6, 7, 8])
    indices = tf.constant([[4], [3], [1] ,[7]])
    updates = tf.constant([9, 10, 11, 12])
    v.scatter_nd_update(indices, updates)
    print(v)

The resulting update to v would look like this:

[1, 11, 3, 10, 9, 6, 7, 12]

See tf.scatter_nd for more details about how to make updates to slices.

scatter_sub

View source

scatter_sub(
    sparse_delta, use_locking=False, name=None
)

Subtracts tf.IndexedSlices from this variable.

scatter_update

View source

scatter_update(
    sparse_delta, use_locking=False, name=None
)

Assigns tf.IndexedSlices to this variable.

set_shape

View source

set_shape(
    shape
)

Overrides the shape for this variable.

sparse_read

View source

sparse_read(
    indices, name=None
)

Gather slices from params axis axis according to indices.

This function supports a subset of tf.gather, see tf.gather for details on usage.

to_proto

View source

to_proto(
    export_scope=None
)

Converts a Variable to a VariableDef protocol buffer.

value

View source

value()

Returns the last snapshot of this variable.

You usually do not need to call this method as all ops that need the value of the variable call it automatically through a convert_to_tensor() call.

Returns a Tensor which holds the value of the variable. You can not assign a new value to this tensor as it is not a reference to the variable.

To avoid copies, if the consumer of the returned value is on the same device as the variable, this actually returns the live value of the variable, not a copy. Updates to the variable are seen by the consumer. If the consumer is on a different device it will get a copy of the variable.

__abs__

View source

__abs__(
    name=None
)

Computes the absolute value of a tensor.

Given a tensor of integer or floating-point values, this operation returns a tensor of the same type, where each element contains the absolute value of the corresponding element in the input.

Given a tensor x of complex numbers, this operation returns a tensor of type float32 or float64 that is the absolute value of each element in x. For a complex number \(a + bj\), its absolute value is computed as \(\sqrt{a^2 + b^2}\).

For example:

# real number
x = tf.constant([-2.25, 3.25])
tf.abs(x)
<tf.Tensor: shape=(2,), dtype=float32,
numpy=array([2.25, 3.25], dtype=float32)>

# complex number
x = tf.constant([[-2.25 + 4.75j], [-3.25 + 5.75j]])
tf.abs(x)
<tf.Tensor: shape=(2, 1), dtype=float64, numpy=
array([[5.25594901],
       [6.60492241]])>

__add__

View source

__add__(
    y
)

The operation invoked by the Tensor.add operator.

Purpose in the API:

This method is exposed in TensorFlow's API so that library developers
can register dispatching for <a href="../tf/Tensor#__add__"><code>Tensor.__add__</code></a> to allow it to handle
custom composite tensors & other custom objects.

The API symbol is not intended to be called by users directly and does
appear in TensorFlow's generated documentation.

__and__

View source

__and__(
    y
)

__div__

View source

__div__(
    y
)

Divides x / y elementwise (using Python 2 division operator semantics). (deprecated)

This function divides x and y, forcing Python 2 semantics. That is, if x and y are both integers then the result will be an integer. This is in contrast to Python 3, where division with / is always a float while division with // is always an integer.

__eq__

View source

__eq__(
    other
)

Compares two variables element-wise for equality.

__floordiv__

View source

__floordiv__(
    y
)

Divides x / y elementwise, rounding toward the most negative integer.

The same as tf.compat.v1.div(x,y) for integers, but uses tf.floor(tf.compat.v1.div(x,y)) for floating point arguments so that the result is always an integer (though possibly an integer represented as floating point). This op is generated by x // y floor division in Python 3 and in Python 2.7 with from __future__ import division.

x and y must have the same type, and the result will have the same type as well.

__ge__

__ge__(
    y, name=None
)

Returns the truth value of (x >= y) element-wise.

Example:

x = tf.constant([5, 4, 6, 7])
y = tf.constant([5, 2, 5, 10])
tf.math.greater_equal(x, y) ==> [True, True, True, False]

x = tf.constant([5, 4, 6, 7])
y = tf.constant([5])
tf.math.greater_equal(x, y) ==> [True, False, True, True]

__getitem__

View source

__getitem__(
    slice_spec
)

Creates a slice helper object given a variable.

This allows creating a sub-tensor from part of the current contents of a variable. See tf.Tensor.getitem for detailed examples of slicing.

This function in addition also allows assignment to a sliced range. This is similar to __setitem__ functionality in Python. However, the syntax is different so that the user can capture the assignment operation for grouping or passing to sess.run(). For example,

import tensorflow as tf
A = tf.Variable([[1,2,3], [4,5,6], [7,8,9]], dtype=tf.float32)
with tf.compat.v1.Session() as sess:
  sess.run(tf.compat.v1.global_variables_initializer())
  print(sess.run(A[:2, :2]))  # => [[1,2], [4,5]]

  op = A[:2,:2].assign(22. * tf.ones((2, 2)))
  print(sess.run(op))  # => [[22, 22, 3], [22, 22, 6], [7,8,9]]

Note that assignments currently do not support NumPy broadcasting semantics.

__gt__

__gt__(
    y, name=None
)

Returns the truth value of (x > y) element-wise.

Example:

x = tf.constant([5, 4, 6])
y = tf.constant([5, 2, 5])
tf.math.greater(x, y) ==> [False, True, True]

x = tf.constant([5, 4, 6])
y = tf.constant([5])
tf.math.greater(x, y) ==> [False, False, True]

__invert__

View source

__invert__(
    name=None
)

__iter__

View source

__iter__()

When executing eagerly, iterates over the value of the variable.

__le__

__le__(
    y, name=None
)

Returns the truth value of (x <= y) element-wise.

Example:

x = tf.constant([5, 4, 6])
y = tf.constant([5])
tf.math.less_equal(x, y) ==> [True, True, False]

x = tf.constant([5, 4, 6])
y = tf.constant([5, 6, 6])
tf.math.less_equal(x, y) ==> [True, True, True]

__lt__

__lt__(
    y, name=None
)

Returns the truth value of (x < y) element-wise.

Example:

x = tf.constant([5, 4, 6])
y = tf.constant([5])
tf.math.less(x, y) ==> [False, True, False]

x = tf.constant([5, 4, 6])
y = tf.constant([5, 6, 7])
tf.math.less(x, y) ==> [False, True, True]

__matmul__

View source

__matmul__(
    y
)

Multiplies matrix a by matrix b, producing a * b.

The inputs must, following any transpositions, be tensors of rank >= 2 where the inner 2 dimensions specify valid matrix multiplication dimensions, and any further outer dimensions specify matching batch size.

Both matrices must be of the same type. The supported types are: bfloat16, float16, float32, float64, int32, int64, complex64, complex128.

Either matrix can be transposed or adjointed (conjugated and transposed) on the fly by setting one of the corresponding flag to True. These are False by default.

If one or both of the matrices contain a lot of zeros, a more efficient multiplication algorithm can be used by setting the corresponding a_is_sparse or b_is_sparse flag to True. These are False by default. This optimization is only available for plain matrices (rank-2 tensors) with datatypes bfloat16 or float32.

A simple 2-D tensor matrix multiplication:

a = tf.constant([1, 2, 3, 4, 5, 6], shape=[2, 3])
a  # 2-D tensor
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[1, 2, 3],
       [4, 5, 6]], dtype=int32)>
b = tf.constant([7, 8, 9, 10, 11, 12], shape=[3, 2])
b  # 2-D tensor
<tf.Tensor: shape=(3, 2), dtype=int32, numpy=
array([[ 7,  8],
       [ 9, 10],
       [11, 12]], dtype=int32)>
c = tf.matmul(a, b)
c  # `a` * `b`
<tf.Tensor: shape=(2, 2), dtype=int32, numpy=
array([[ 58,  64],
       [139, 154]], dtype=int32)>

A batch matrix multiplication with batch shape [2]:

a = tf.constant(np.arange(1, 13, dtype=np.int32), shape=[2, 2, 3])
a  # 3-D tensor
<tf.Tensor: shape=(2, 2, 3), dtype=int32, numpy=
array([[[ 1,  2,  3],
        [ 4,  5,  6]],
       [[ 7,  8,  9],
        [10, 11, 12]]], dtype=int32)>
b = tf.constant(np.arange(13, 25, dtype=np.int32), shape=[2, 3, 2])
b  # 3-D tensor
<tf.Tensor: shape=(2, 3, 2), dtype=int32, numpy=
array([[[13, 14],
        [15, 16],
        [17, 18]],
       [[19, 20],
        [21, 22],
        [23, 24]]], dtype=int32)>
c = tf.matmul(a, b)
c  # `a` * `b`
<tf.Tensor: shape=(2, 2, 2), dtype=int32, numpy=
array([[[ 94, 100],
        [229, 244]],
       [[508, 532],
        [697, 730]]], dtype=int32)>

Since python >= 3.5 the @ operator is supported (see PEP 465). In TensorFlow, it simply calls the tf.matmul() function, so the following lines are equivalent:

d = a @ b @ [[10], [11]]
d = tf.matmul(tf.matmul(a, b), [[10], [11]])

__mod__

View source

__mod__(
    y
)

Returns element-wise remainder of division. When x < 0 xor y < 0 is

true, this follows Python semantics in that the result here is consistent with a flooring divide. E.g. floor(x / y) * y + mod(x, y) = x.

__mul__

View source

__mul__(
    y
)

Dispatches cwise mul for "DenseDense" and "DenseSparse".

__ne__

View source

__ne__(
    other
)

Compares two variables element-wise for equality.

__neg__

__neg__(
    name=None
)

Computes numerical negative value element-wise.

I.e., \(y = -x\).

__or__

View source

__or__(
    y
)

__pow__

View source

__pow__(
    y
)

Computes the power of one value to another.

Given a tensor x and a tensor y, this operation computes \(x^y\) for corresponding elements in x and y. For example:

x = tf.constant([[2, 2], [3, 3]])
y = tf.constant([[8, 16], [2, 3]])
tf.pow(x, y)  # [[256, 65536], [9, 27]]

__radd__

View source

__radd__(
    x
)

The operation invoked by the Tensor.add operator.

Purpose in the API:

This method is exposed in TensorFlow's API so that library developers
can register dispatching for <a href="../tf/Tensor#__add__"><code>Tensor.__add__</code></a> to allow it to handle
custom composite tensors & other custom objects.

The API symbol is not intended to be called by users directly and does
appear in TensorFlow's generated documentation.

__rand__

View source

__rand__(
    x
)

__rdiv__

View source

__rdiv__(
    x
)

Divides x / y elementwise (using Python 2 division operator semantics). (deprecated)

This function divides x and y, forcing Python 2 semantics. That is, if x and y are both integers then the result will be an integer. This is in contrast to Python 3, where division with / is always a float while division with // is always an integer.

__rfloordiv__

View source

__rfloordiv__(
    x
)

Divides x / y elementwise, rounding toward the most negative integer.

The same as tf.compat.v1.div(x,y) for integers, but uses tf.floor(tf.compat.v1.div(x,y)) for floating point arguments so that the result is always an integer (though possibly an integer represented as floating point). This op is generated by x // y floor division in Python 3 and in Python 2.7 with from __future__ import division.

x and y must have the same type, and the result will have the same type as well.

__rmatmul__

View source

__rmatmul__(
    x
)

Multiplies matrix a by matrix b, producing a * b.

The inputs must, following any transpositions, be tensors of rank >= 2 where the inner 2 dimensions specify valid matrix multiplication dimensions, and any further outer dimensions specify matching batch size.

Both matrices must be of the same type. The supported types are: bfloat16, float16, float32, float64, int32, int64, complex64, complex128.

Either matrix can be transposed or adjointed (conjugated and transposed) on the fly by setting one of the corresponding flag to True. These are False by default.

If one or both of the matrices contain a lot of zeros, a more efficient multiplication algorithm can be used by setting the corresponding a_is_sparse or b_is_sparse flag to True. These are False by default. This optimization is only available for plain matrices (rank-2 tensors) with datatypes bfloat16 or float32.

A simple 2-D tensor matrix multiplication:

a = tf.constant([1, 2, 3, 4, 5, 6], shape=[2, 3])
a  # 2-D tensor
<tf.Tensor: shape=(2, 3), dtype=int32, numpy=
array([[1, 2, 3],
       [4, 5, 6]], dtype=int32)>
b = tf.constant([7, 8, 9, 10, 11, 12], shape=[3, 2])
b  # 2-D tensor
<tf.Tensor: shape=(3, 2), dtype=int32, numpy=
array([[ 7,  8],
       [ 9, 10],
       [11, 12]], dtype=int32)>
c = tf.matmul(a, b)
c  # `a` * `b`
<tf.Tensor: shape=(2, 2), dtype=int32, numpy=
array([[ 58,  64],
       [139, 154]], dtype=int32)>

A batch matrix multiplication with batch shape [2]:

a = tf.constant(np.arange(1, 13, dtype=np.int32), shape=[2, 2, 3])
a  # 3-D tensor
<tf.Tensor: shape=(2, 2, 3), dtype=int32, numpy=
array([[[ 1,  2,  3],
        [ 4,  5,  6]],
       [[ 7,  8,  9],
        [10, 11, 12]]], dtype=int32)>
b = tf.constant(np.arange(13, 25, dtype=np.int32), shape=[2, 3, 2])
b  # 3-D tensor
<tf.Tensor: shape=(2, 3, 2), dtype=int32, numpy=
array([[[13, 14],
        [15, 16],
        [17, 18]],
       [[19, 20],
        [21, 22],
        [23, 24]]], dtype=int32)>
c = tf.matmul(a, b)
c  # `a` * `b`
<tf.Tensor: shape=(2, 2, 2), dtype=int32, numpy=
array([[[ 94, 100],
        [229, 244]],
       [[508, 532],
        [697, 730]]], dtype=int32)>

Since python >= 3.5 the @ operator is supported (see PEP 465). In TensorFlow, it simply calls the tf.matmul() function, so the following lines are equivalent:

d = a @ b @ [[10], [11]]
d = tf.matmul(tf.matmul(a, b), [[10], [11]])

__rmod__

View source

__rmod__(
    x
)

Returns element-wise remainder of division. When x < 0 xor y < 0 is

true, this follows Python semantics in that the result here is consistent with a flooring divide. E.g. floor(x / y) * y + mod(x, y) = x.

__rmul__

View source

__rmul__(
    x
)

Dispatches cwise mul for "DenseDense" and "DenseSparse".

__ror__

View source

__ror__(
    x
)

__rpow__

View source

__rpow__(
    x
)

Computes the power of one value to another.

Given a tensor x and a tensor y, this operation computes \(x^y\) for corresponding elements in x and y. For example:

x = tf.constant([[2, 2], [3, 3]])
y = tf.constant([[8, 16], [2, 3]])
tf.pow(x, y)  # [[256, 65536], [9, 27]]

__rsub__

View source

__rsub__(
    x
)

Returns x - y element-wise.

Both input and output have a range (-inf, inf).

Example usages below.

Subtract operation between an array and a scalar:

x = [1, 2, 3, 4, 5]
y = 1
tf.subtract(x, y)
<tf.Tensor: shape=(5,), dtype=int32, numpy=array([0, 1, 2, 3, 4], dtype=int32)>
tf.subtract(y, x)
<tf.Tensor: shape=(5,), dtype=int32,
numpy=array([ 0, -1, -2, -3, -4], dtype=int32)>

Note that binary - operator can be used instead:

x = tf.convert_to_tensor([1, 2, 3, 4, 5])
y = tf.convert_to_tensor(1)
x - y
<tf.Tensor: shape=(5,), dtype=int32, numpy=array([0, 1, 2, 3, 4], dtype=int32)>

Subtract operation between an array and a tensor of same shape:

x = [1, 2, 3, 4, 5]
y = tf.constant([5, 4, 3, 2, 1])
tf.subtract(y, x)
<tf.Tensor: shape=(5,), dtype=int32,
numpy=array([ 4,  2,  0, -2, -4], dtype=int32)>

For example,

x = tf.constant([1, 2], dtype=tf.int8)
y = [2**8 + 1, 2**8 + 2]
tf.subtract(x, y)
<tf.Tensor: shape=(2,), dtype=int8, numpy=array([0, 0], dtype=int8)>

When subtracting two input values of different shapes, tf.subtract follows the general broadcasting rules . The two input array shapes are compared element-wise. Starting with the trailing dimensions, the two dimensions either have to be equal or one of them needs to be 1.

For example,

x = np.ones(6).reshape(2, 3, 1)
y = np.ones(6).reshape(2, 1, 3)
tf.subtract(x, y)
<tf.Tensor: shape=(2, 3, 3), dtype=float64, numpy=
array([[[0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]],
       [[0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]]])>

Example with inputs of different dimensions:

x = np.ones(6).reshape(2, 3, 1)
y = np.ones(6).reshape(1, 6)
tf.subtract(x, y)
<tf.Tensor: shape=(2, 3, 6), dtype=float64, numpy=
array([[[0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0.]],
       [[0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0.]]])>

__rtruediv__

View source

__rtruediv__(
    x
)

Divides x / y elementwise (using Python 3 division operator semantics).

This function forces Python 3 division operator semantics where all integer arguments are cast to floating types first. This op is generated by normal x / y division in Python 3 and in Python 2.7 with from __future__ import division. If you want integer division that rounds down, use x // y or tf.math.floordiv.

x and y must have the same numeric type. If the inputs are floating point, the output will have the same type. If the inputs are integral, the inputs are cast to float32 for int8 and int16 and float64 for int32 and int64 (matching the behavior of Numpy).

__rxor__

View source

__rxor__(
    x
)

__sub__

View source

__sub__(
    y
)

Returns x - y element-wise.

Both input and output have a range (-inf, inf).

Example usages below.

Subtract operation between an array and a scalar:

x = [1, 2, 3, 4, 5]
y = 1
tf.subtract(x, y)
<tf.Tensor: shape=(5,), dtype=int32, numpy=array([0, 1, 2, 3, 4], dtype=int32)>
tf.subtract(y, x)
<tf.Tensor: shape=(5,), dtype=int32,
numpy=array([ 0, -1, -2, -3, -4], dtype=int32)>

Note that binary - operator can be used instead:

x = tf.convert_to_tensor([1, 2, 3, 4, 5])
y = tf.convert_to_tensor(1)
x - y
<tf.Tensor: shape=(5,), dtype=int32, numpy=array([0, 1, 2, 3, 4], dtype=int32)>

Subtract operation between an array and a tensor of same shape:

x = [1, 2, 3, 4, 5]
y = tf.constant([5, 4, 3, 2, 1])
tf.subtract(y, x)
<tf.Tensor: shape=(5,), dtype=int32,
numpy=array([ 4,  2,  0, -2, -4], dtype=int32)>

For example,

x = tf.constant([1, 2], dtype=tf.int8)
y = [2**8 + 1, 2**8 + 2]
tf.subtract(x, y)
<tf.Tensor: shape=(2,), dtype=int8, numpy=array([0, 0], dtype=int8)>

When subtracting two input values of different shapes, tf.subtract follows the general broadcasting rules . The two input array shapes are compared element-wise. Starting with the trailing dimensions, the two dimensions either have to be equal or one of them needs to be 1.

For example,

x = np.ones(6).reshape(2, 3, 1)
y = np.ones(6).reshape(2, 1, 3)
tf.subtract(x, y)
<tf.Tensor: shape=(2, 3, 3), dtype=float64, numpy=
array([[[0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]],
       [[0., 0., 0.],
        [0., 0., 0.],
        [0., 0., 0.]]])>

Example with inputs of different dimensions:

x = np.ones(6).reshape(2, 3, 1)
y = np.ones(6).reshape(1, 6)
tf.subtract(x, y)
<tf.Tensor: shape=(2, 3, 6), dtype=float64, numpy=
array([[[0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0.]],
       [[0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0.]]])>