View source on GitHub |
Represents the type of the elements in a Tensor
.
Inherits From: TraceType
tf.dtypes.DType()
DType
's are used to specify the output data type for operations which
require it, or to inspect the data type of existing Tensor
's.
Examples:
tf.constant(1, dtype=tf.int64)
<tf.Tensor: shape=(), dtype=int64, numpy=1>
tf.constant(1.0).dtype
tf.float32
See tf.dtypes
for a complete list of DType
's defined.
Methods
experimental_as_proto
experimental_as_proto() -> types_pb2.SerializedDType
Returns a proto representation of the Dtype instance.
experimental_from_proto
@classmethod
experimental_from_proto( proto: types_pb2.SerializedDType ) -> 'DType'
Returns a Dtype instance based on the serialized proto.
experimental_type_proto
@classmethod
experimental_type_proto() -> Type[types_pb2.SerializedDType]
Returns the type of proto associated with DType serialization.
is_compatible_with
is_compatible_with(
other
)
Returns True if the other
DType will be converted to this DType.
The conversion rules are as follows:
DType(T) .is_compatible_with(DType(T)) == True
Args | |
---|---|
other
|
A DType (or object that may be converted to a DType ).
|
Returns | |
---|---|
True if a Tensor of the other DType will be implicitly converted to
this DType .
|
is_subtype_of
is_subtype_of(
other: tf.types.experimental.TraceType
) -> bool
See tf.types.experimental.TraceType base class.
most_specific_common_supertype
most_specific_common_supertype(
types: Sequence[tf.types.experimental.TraceType
]
) -> Optional['DType']
See tf.types.experimental.TraceType base class.
__eq__
__eq__(
other
)
Returns True iff this DType refers to the same type as other
.
__ne__
__ne__(
other
)
Returns True iff self != other.