Class BoundedTensorSpec
Inherits From: TensorSpec
Defined in tensorflow/python/framework/tensor_spec.py
.
A TensorSpec
that specifies minimum and maximum values.
Example usage:
spec = tensor_spec.BoundedTensorSpec((1, 2, 3), tf.float32, 0, (5, 5, 5))
tf_minimum = tf.convert_to_tensor(spec.minimum, dtype=spec.dtype)
tf_maximum = tf.convert_to_tensor(spec.maximum, dtype=spec.dtype)
Bounds are meant to be inclusive. This is especially important for integer types. The following spec will be satisfied by tensors with values in the set {0, 1, 2}:
spec = tensor_spec.BoundedTensorSpec((3, 5), tf.int32, 0, 2)
__init__
__init__(
shape,
dtype,
minimum,
maximum,
name=None
)
Initializes a new BoundedTensorSpec
.
Args:
shape
: Value convertible totf.TensorShape
. The shape of the tensor.dtype
: Value convertible totf.DType
. The type of the tensor values.minimum
: Number or sequence specifying the minimum element bounds (inclusive). Must be broadcastable toshape
.maximum
: Number or sequence specifying the maximum element bounds (inclusive). Must be broadcastable toshape
.name
: Optional string containing a semantic name for the corresponding array. Defaults toNone
.
Raises:
ValueError
: Ifminimum
ormaximum
are not provided or not broadcastable toshape
.TypeError
: If the shape is not an iterable or if thedtype
is an invalid numpy dtype.
Properties
dtype
Returns the dtype
of elements in the tensor.
maximum
Returns a NumPy array specifying the maximum bounds (inclusive).
minimum
Returns a NumPy array specifying the minimum bounds (inclusive).
name
Returns the (optionally provided) name of the described tensor.
shape
Returns the TensorShape
that represents the shape of the tensor.
Methods
tf.contrib.framework.BoundedTensorSpec.__eq__
__eq__(other)
Return self==value.
tf.contrib.framework.BoundedTensorSpec.__ne__
__ne__(other)
Return self!=value.
tf.contrib.framework.BoundedTensorSpec.from_spec
@classmethod
from_spec(
cls,
spec
)
tf.contrib.framework.BoundedTensorSpec.from_tensor
from_tensor(
cls,
tensor,
name=None
)
tf.contrib.framework.BoundedTensorSpec.is_compatible_with
is_compatible_with(spec_or_tensor)
Returns True if spec_or_tensor is compatible with this TensorSpec.
Two tensors are considered compatible if they have the same dtype
and their shapes are compatible (see tf.TensorShape.is_compatible_with
).
Args:
spec_or_tensor
: A tf.TensorSpec or a tf.Tensor
Returns:
True if spec_or_tensor is compatible with self.