Go: Update generated wrapper functions for TensorFlow ops.
PiperOrigin-RevId: 249541834
This commit is contained in:
A. Unique TensorFlower
TensorFlower Gardener
parent
e070fe045d
commit
4f4921ce0b
@ -5587,6 +5587,78 @@ func MapSize(scope *Scope, dtypes []tf.DataType, optional ...MapSizeAttr) (size
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// MapUnstageNoKeyAttr is an optional argument to MapUnstageNoKey.
|
||||
type MapUnstageNoKeyAttr func(optionalAttr)
|
||||
|
||||
// MapUnstageNoKeyCapacity sets the optional capacity attribute to value.
|
||||
// If not specified, defaults to 0
|
||||
//
|
||||
// REQUIRES: value >= 0
|
||||
func MapUnstageNoKeyCapacity(value int64) MapUnstageNoKeyAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["capacity"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MapUnstageNoKeyMemoryLimit sets the optional memory_limit attribute to value.
|
||||
// If not specified, defaults to 0
|
||||
//
|
||||
// REQUIRES: value >= 0
|
||||
func MapUnstageNoKeyMemoryLimit(value int64) MapUnstageNoKeyAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["memory_limit"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MapUnstageNoKeyContainer sets the optional container attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func MapUnstageNoKeyContainer(value string) MapUnstageNoKeyAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["container"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MapUnstageNoKeySharedName sets the optional shared_name attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func MapUnstageNoKeySharedName(value string) MapUnstageNoKeyAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["shared_name"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Op removes and returns a random (key, value)
|
||||
//
|
||||
// from the underlying container. If the underlying container
|
||||
// does not contain elements, the op will block until it does.
|
||||
func MapUnstageNoKey(scope *Scope, indices tf.Output, dtypes []tf.DataType, optional ...MapUnstageNoKeyAttr) (key tf.Output, values []tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"dtypes": dtypes}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "MapUnstageNoKey",
|
||||
Input: []tf.Input{
|
||||
indices,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
var idx int
|
||||
var err error
|
||||
key = op.Output(idx)
|
||||
if values, idx, err = makeOutputList(op, idx, "values"); err != nil {
|
||||
scope.UpdateErr("MapUnstageNoKey", err)
|
||||
return
|
||||
}
|
||||
return key, values
|
||||
}
|
||||
|
||||
// MapUnstageAttr is an optional argument to MapUnstage.
|
||||
type MapUnstageAttr func(optionalAttr)
|
||||
|
||||
@ -18535,76 +18607,348 @@ func ReduceJoin(scope *Scope, inputs tf.Output, reduction_indices tf.Output, opt
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// MapUnstageNoKeyAttr is an optional argument to MapUnstageNoKey.
|
||||
type MapUnstageNoKeyAttr func(optionalAttr)
|
||||
// MutableDenseHashTableV2Attr is an optional argument to MutableDenseHashTableV2.
|
||||
type MutableDenseHashTableV2Attr func(optionalAttr)
|
||||
|
||||
// MapUnstageNoKeyCapacity sets the optional capacity attribute to value.
|
||||
// If not specified, defaults to 0
|
||||
// MutableDenseHashTableV2Container sets the optional container attribute to value.
|
||||
//
|
||||
// REQUIRES: value >= 0
|
||||
func MapUnstageNoKeyCapacity(value int64) MapUnstageNoKeyAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["capacity"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MapUnstageNoKeyMemoryLimit sets the optional memory_limit attribute to value.
|
||||
// If not specified, defaults to 0
|
||||
//
|
||||
// REQUIRES: value >= 0
|
||||
func MapUnstageNoKeyMemoryLimit(value int64) MapUnstageNoKeyAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["memory_limit"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MapUnstageNoKeyContainer sets the optional container attribute to value.
|
||||
// value: If non-empty, this table is placed in the given container.
|
||||
// Otherwise, a default container is used.
|
||||
// If not specified, defaults to ""
|
||||
func MapUnstageNoKeyContainer(value string) MapUnstageNoKeyAttr {
|
||||
func MutableDenseHashTableV2Container(value string) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["container"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MapUnstageNoKeySharedName sets the optional shared_name attribute to value.
|
||||
// MutableDenseHashTableV2SharedName sets the optional shared_name attribute to value.
|
||||
//
|
||||
// value: If non-empty, this table is shared under the given name across
|
||||
// multiple sessions.
|
||||
// If not specified, defaults to ""
|
||||
func MapUnstageNoKeySharedName(value string) MapUnstageNoKeyAttr {
|
||||
func MutableDenseHashTableV2SharedName(value string) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["shared_name"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Op removes and returns a random (key, value)
|
||||
// MutableDenseHashTableV2UseNodeNameSharing sets the optional use_node_name_sharing attribute to value.
|
||||
// If not specified, defaults to false
|
||||
func MutableDenseHashTableV2UseNodeNameSharing(value bool) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["use_node_name_sharing"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2ValueShape sets the optional value_shape attribute to value.
|
||||
//
|
||||
// from the underlying container. If the underlying container
|
||||
// does not contain elements, the op will block until it does.
|
||||
func MapUnstageNoKey(scope *Scope, indices tf.Output, dtypes []tf.DataType, optional ...MapUnstageNoKeyAttr) (key tf.Output, values []tf.Output) {
|
||||
// value: The shape of each value.
|
||||
// If not specified, defaults to <>
|
||||
func MutableDenseHashTableV2ValueShape(value tf.Shape) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["value_shape"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2InitialNumBuckets sets the optional initial_num_buckets attribute to value.
|
||||
//
|
||||
// value: The initial number of hash table buckets. Must be a power
|
||||
// to 2.
|
||||
// If not specified, defaults to 131072
|
||||
func MutableDenseHashTableV2InitialNumBuckets(value int64) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["initial_num_buckets"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2MaxLoadFactor sets the optional max_load_factor attribute to value.
|
||||
//
|
||||
// value: The maximum ratio between number of entries and number of
|
||||
// buckets before growing the table. Must be between 0 and 1.
|
||||
// If not specified, defaults to 0.8
|
||||
func MutableDenseHashTableV2MaxLoadFactor(value float32) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["max_load_factor"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Creates an empty hash table that uses tensors as the backing store.
|
||||
//
|
||||
// It uses "open addressing" with quadratic reprobing to resolve
|
||||
// collisions.
|
||||
//
|
||||
// This op creates a mutable hash table, specifying the type of its keys and
|
||||
// values. Each value must be a scalar. Data can be inserted into the table using
|
||||
// the insert operations. It does not support the initialization operation.
|
||||
//
|
||||
// Arguments:
|
||||
// empty_key: The key used to represent empty key buckets internally. Must not
|
||||
// be used in insert or lookup operations.
|
||||
//
|
||||
// value_dtype: Type of the table values.
|
||||
//
|
||||
// Returns Handle to a table.
|
||||
func MutableDenseHashTableV2(scope *Scope, empty_key tf.Output, deleted_key tf.Output, value_dtype tf.DataType, optional ...MutableDenseHashTableV2Attr) (table_handle tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"dtypes": dtypes}
|
||||
attrs := map[string]interface{}{"value_dtype": value_dtype}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "MapUnstageNoKey",
|
||||
Type: "MutableDenseHashTableV2",
|
||||
Input: []tf.Input{
|
||||
indices,
|
||||
empty_key, deleted_key,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// Returns the element-wise max of two SparseTensors.
|
||||
//
|
||||
// Assumes the two SparseTensors have the same shape, i.e., no broadcasting.
|
||||
//
|
||||
// Arguments:
|
||||
// a_indices: 2-D. `N x R` matrix with the indices of non-empty values in a
|
||||
// SparseTensor, in the canonical lexicographic ordering.
|
||||
// a_values: 1-D. `N` non-empty values corresponding to `a_indices`.
|
||||
// a_shape: 1-D. Shape of the input SparseTensor.
|
||||
// b_indices: counterpart to `a_indices` for the other operand.
|
||||
// b_values: counterpart to `a_values` for the other operand; must be of the same dtype.
|
||||
// b_shape: counterpart to `a_shape` for the other operand; the two shapes must be equal.
|
||||
//
|
||||
// Returns 2-D. The indices of the output SparseTensor.1-D. The values of the output SparseTensor.
|
||||
func SparseSparseMaximum(scope *Scope, a_indices tf.Output, a_values tf.Output, a_shape tf.Output, b_indices tf.Output, b_values tf.Output, b_shape tf.Output) (output_indices tf.Output, output_values tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
var idx int
|
||||
var err error
|
||||
key = op.Output(idx)
|
||||
if values, idx, err = makeOutputList(op, idx, "values"); err != nil {
|
||||
scope.UpdateErr("MapUnstageNoKey", err)
|
||||
opspec := tf.OpSpec{
|
||||
Type: "SparseSparseMaximum",
|
||||
Input: []tf.Input{
|
||||
a_indices, a_values, a_shape, b_indices, b_values, b_shape,
|
||||
},
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0), op.Output(1)
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataAttr is an optional argument to TPUReplicateMetadata.
|
||||
type TPUReplicateMetadataAttr func(optionalAttr)
|
||||
|
||||
// TPUReplicateMetadataNumCoresPerReplica sets the optional num_cores_per_replica attribute to value.
|
||||
//
|
||||
// value: Number of cores per replica. Used for model parallelism.
|
||||
// If not specified, defaults to 1
|
||||
func TPUReplicateMetadataNumCoresPerReplica(value int64) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["num_cores_per_replica"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataTopology sets the optional topology attribute to value.
|
||||
//
|
||||
// value: TopologyProto indicating the topology of the TPU pod slice.
|
||||
// If not specified, defaults to ""
|
||||
func TPUReplicateMetadataTopology(value string) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["topology"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataUseTpu sets the optional use_tpu attribute to value.
|
||||
//
|
||||
// value: Whether to place the computation on the TPU.
|
||||
// If not specified, defaults to true
|
||||
func TPUReplicateMetadataUseTpu(value bool) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["use_tpu"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataDeviceAssignment sets the optional device_assignment attribute to value.
|
||||
//
|
||||
// value: The assignment of devices for the computation.
|
||||
// If not specified, defaults to <>
|
||||
func TPUReplicateMetadataDeviceAssignment(value []int64) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["device_assignment"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataComputationShape sets the optional computation_shape attribute to value.
|
||||
//
|
||||
// value: DEPRECATED. Use num_cores_per_replica instead.
|
||||
// If not specified, defaults to <>
|
||||
func TPUReplicateMetadataComputationShape(value []int64) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["computation_shape"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataHostComputeCore sets the optional host_compute_core attribute to value.
|
||||
// If not specified, defaults to <>
|
||||
func TPUReplicateMetadataHostComputeCore(value []string) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["host_compute_core"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataPaddingMap sets the optional padding_map attribute to value.
|
||||
// If not specified, defaults to <>
|
||||
func TPUReplicateMetadataPaddingMap(value []string) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["padding_map"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataStepMarkerLocation sets the optional step_marker_location attribute to value.
|
||||
// If not specified, defaults to "STEP_MARK_AT_ENTRY"
|
||||
func TPUReplicateMetadataStepMarkerLocation(value string) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["step_marker_location"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Metadata indicaitng how the TPU computation should be replicated.
|
||||
//
|
||||
// Arguments:
|
||||
// num_replicas: Number of replicas of the computation
|
||||
//
|
||||
// Returns the created operation.
|
||||
func TPUReplicateMetadata(scope *Scope, num_replicas int64, optional ...TPUReplicateMetadataAttr) (o *tf.Operation) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
return key, values
|
||||
attrs := map[string]interface{}{"num_replicas": num_replicas}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "TPUReplicateMetadata",
|
||||
|
||||
Attrs: attrs,
|
||||
}
|
||||
return scope.AddOperation(opspec)
|
||||
}
|
||||
|
||||
// Fast Fourier transform.
|
||||
//
|
||||
// Computes the 1-dimensional discrete Fourier transform over the inner-most
|
||||
// dimension of `input`.
|
||||
//
|
||||
// Arguments:
|
||||
// input: A complex tensor.
|
||||
//
|
||||
// Returns A complex tensor of the same shape as `input`. The inner-most
|
||||
// dimension of `input` is replaced with its 1D Fourier transform.
|
||||
//
|
||||
// @compatibility(numpy)
|
||||
// Equivalent to np.fft.fft
|
||||
// @end_compatibility
|
||||
func FFT(scope *Scope, input tf.Output) (output tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "FFT",
|
||||
Input: []tf.Input{
|
||||
input,
|
||||
},
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// MaxAttr is an optional argument to Max.
|
||||
type MaxAttr func(optionalAttr)
|
||||
|
||||
// MaxKeepDims sets the optional keep_dims attribute to value.
|
||||
//
|
||||
// value: If true, retain reduced dimensions with length 1.
|
||||
// If not specified, defaults to false
|
||||
func MaxKeepDims(value bool) MaxAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["keep_dims"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Computes the maximum of elements across dimensions of a tensor.
|
||||
//
|
||||
// Reduces `input` along the dimensions given in `axis`. Unless
|
||||
// `keep_dims` is true, the rank of the tensor is reduced by 1 for each entry in
|
||||
// `axis`. If `keep_dims` is true, the reduced dimensions are
|
||||
// retained with length 1.
|
||||
//
|
||||
// Arguments:
|
||||
// input: The tensor to reduce.
|
||||
// axis: The dimensions to reduce. Must be in the range
|
||||
// `[-rank(input), rank(input))`.
|
||||
//
|
||||
// Returns The reduced tensor.
|
||||
func Max(scope *Scope, input tf.Output, axis tf.Output, optional ...MaxAttr) (output tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "Max",
|
||||
Input: []tf.Input{
|
||||
input, axis,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// StatelessRandomUniformAttr is an optional argument to StatelessRandomUniform.
|
||||
type StatelessRandomUniformAttr func(optionalAttr)
|
||||
|
||||
// StatelessRandomUniformDtype sets the optional dtype attribute to value.
|
||||
//
|
||||
// value: The type of the output.
|
||||
// If not specified, defaults to DT_FLOAT
|
||||
func StatelessRandomUniformDtype(value tf.DataType) StatelessRandomUniformAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["dtype"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Outputs deterministic pseudorandom random values from a uniform distribution.
|
||||
//
|
||||
// The generated values follow a uniform distribution in the range `[0, 1)`. The
|
||||
// lower bound 0 is included in the range, while the upper bound 1 is excluded.
|
||||
//
|
||||
// The outputs are a deterministic function of `shape` and `seed`.
|
||||
//
|
||||
// Arguments:
|
||||
// shape: The shape of the output tensor.
|
||||
// seed: 2 seeds (shape [2]).
|
||||
//
|
||||
// Returns Random values with specified shape.
|
||||
func StatelessRandomUniform(scope *Scope, shape tf.Output, seed tf.Output, optional ...StatelessRandomUniformAttr) (output tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "StatelessRandomUniform",
|
||||
Input: []tf.Input{
|
||||
shape, seed,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// RetrieveTPUEmbeddingAdagradParametersGradAccumDebugAttr is an optional argument to RetrieveTPUEmbeddingAdagradParametersGradAccumDebug.
|
||||
@ -19351,250 +19695,6 @@ func RandomGammaGrad(scope *Scope, alpha tf.Output, sample tf.Output) (output tf
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// Returns the element-wise max of two SparseTensors.
|
||||
//
|
||||
// Assumes the two SparseTensors have the same shape, i.e., no broadcasting.
|
||||
//
|
||||
// Arguments:
|
||||
// a_indices: 2-D. `N x R` matrix with the indices of non-empty values in a
|
||||
// SparseTensor, in the canonical lexicographic ordering.
|
||||
// a_values: 1-D. `N` non-empty values corresponding to `a_indices`.
|
||||
// a_shape: 1-D. Shape of the input SparseTensor.
|
||||
// b_indices: counterpart to `a_indices` for the other operand.
|
||||
// b_values: counterpart to `a_values` for the other operand; must be of the same dtype.
|
||||
// b_shape: counterpart to `a_shape` for the other operand; the two shapes must be equal.
|
||||
//
|
||||
// Returns 2-D. The indices of the output SparseTensor.1-D. The values of the output SparseTensor.
|
||||
func SparseSparseMaximum(scope *Scope, a_indices tf.Output, a_values tf.Output, a_shape tf.Output, b_indices tf.Output, b_values tf.Output, b_shape tf.Output) (output_indices tf.Output, output_values tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "SparseSparseMaximum",
|
||||
Input: []tf.Input{
|
||||
a_indices, a_values, a_shape, b_indices, b_values, b_shape,
|
||||
},
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0), op.Output(1)
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataAttr is an optional argument to TPUReplicateMetadata.
|
||||
type TPUReplicateMetadataAttr func(optionalAttr)
|
||||
|
||||
// TPUReplicateMetadataNumCoresPerReplica sets the optional num_cores_per_replica attribute to value.
|
||||
//
|
||||
// value: Number of cores per replica. Used for model parallelism.
|
||||
// If not specified, defaults to 1
|
||||
func TPUReplicateMetadataNumCoresPerReplica(value int64) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["num_cores_per_replica"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataTopology sets the optional topology attribute to value.
|
||||
//
|
||||
// value: TopologyProto indicating the topology of the TPU pod slice.
|
||||
// If not specified, defaults to ""
|
||||
func TPUReplicateMetadataTopology(value string) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["topology"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataUseTpu sets the optional use_tpu attribute to value.
|
||||
//
|
||||
// value: Whether to place the computation on the TPU.
|
||||
// If not specified, defaults to true
|
||||
func TPUReplicateMetadataUseTpu(value bool) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["use_tpu"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataDeviceAssignment sets the optional device_assignment attribute to value.
|
||||
//
|
||||
// value: The assignment of devices for the computation.
|
||||
// If not specified, defaults to <>
|
||||
func TPUReplicateMetadataDeviceAssignment(value []int64) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["device_assignment"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataComputationShape sets the optional computation_shape attribute to value.
|
||||
//
|
||||
// value: DEPRECATED. Use num_cores_per_replica instead.
|
||||
// If not specified, defaults to <>
|
||||
func TPUReplicateMetadataComputationShape(value []int64) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["computation_shape"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataHostComputeCore sets the optional host_compute_core attribute to value.
|
||||
// If not specified, defaults to <>
|
||||
func TPUReplicateMetadataHostComputeCore(value []string) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["host_compute_core"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataPaddingMap sets the optional padding_map attribute to value.
|
||||
// If not specified, defaults to <>
|
||||
func TPUReplicateMetadataPaddingMap(value []string) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["padding_map"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// TPUReplicateMetadataStepMarkerLocation sets the optional step_marker_location attribute to value.
|
||||
// If not specified, defaults to "STEP_MARK_AT_ENTRY"
|
||||
func TPUReplicateMetadataStepMarkerLocation(value string) TPUReplicateMetadataAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["step_marker_location"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Metadata indicaitng how the TPU computation should be replicated.
|
||||
//
|
||||
// Arguments:
|
||||
// num_replicas: Number of replicas of the computation
|
||||
//
|
||||
// Returns the created operation.
|
||||
func TPUReplicateMetadata(scope *Scope, num_replicas int64, optional ...TPUReplicateMetadataAttr) (o *tf.Operation) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"num_replicas": num_replicas}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "TPUReplicateMetadata",
|
||||
|
||||
Attrs: attrs,
|
||||
}
|
||||
return scope.AddOperation(opspec)
|
||||
}
|
||||
|
||||
// Fast Fourier transform.
|
||||
//
|
||||
// Computes the 1-dimensional discrete Fourier transform over the inner-most
|
||||
// dimension of `input`.
|
||||
//
|
||||
// Arguments:
|
||||
// input: A complex tensor.
|
||||
//
|
||||
// Returns A complex tensor of the same shape as `input`. The inner-most
|
||||
// dimension of `input` is replaced with its 1D Fourier transform.
|
||||
//
|
||||
// @compatibility(numpy)
|
||||
// Equivalent to np.fft.fft
|
||||
// @end_compatibility
|
||||
func FFT(scope *Scope, input tf.Output) (output tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "FFT",
|
||||
Input: []tf.Input{
|
||||
input,
|
||||
},
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// MaxAttr is an optional argument to Max.
|
||||
type MaxAttr func(optionalAttr)
|
||||
|
||||
// MaxKeepDims sets the optional keep_dims attribute to value.
|
||||
//
|
||||
// value: If true, retain reduced dimensions with length 1.
|
||||
// If not specified, defaults to false
|
||||
func MaxKeepDims(value bool) MaxAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["keep_dims"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Computes the maximum of elements across dimensions of a tensor.
|
||||
//
|
||||
// Reduces `input` along the dimensions given in `axis`. Unless
|
||||
// `keep_dims` is true, the rank of the tensor is reduced by 1 for each entry in
|
||||
// `axis`. If `keep_dims` is true, the reduced dimensions are
|
||||
// retained with length 1.
|
||||
//
|
||||
// Arguments:
|
||||
// input: The tensor to reduce.
|
||||
// axis: The dimensions to reduce. Must be in the range
|
||||
// `[-rank(input), rank(input))`.
|
||||
//
|
||||
// Returns The reduced tensor.
|
||||
func Max(scope *Scope, input tf.Output, axis tf.Output, optional ...MaxAttr) (output tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "Max",
|
||||
Input: []tf.Input{
|
||||
input, axis,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// StatelessRandomUniformAttr is an optional argument to StatelessRandomUniform.
|
||||
type StatelessRandomUniformAttr func(optionalAttr)
|
||||
|
||||
// StatelessRandomUniformDtype sets the optional dtype attribute to value.
|
||||
//
|
||||
// value: The type of the output.
|
||||
// If not specified, defaults to DT_FLOAT
|
||||
func StatelessRandomUniformDtype(value tf.DataType) StatelessRandomUniformAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["dtype"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Outputs deterministic pseudorandom random values from a uniform distribution.
|
||||
//
|
||||
// The generated values follow a uniform distribution in the range `[0, 1)`. The
|
||||
// lower bound 0 is included in the range, while the upper bound 1 is excluded.
|
||||
//
|
||||
// The outputs are a deterministic function of `shape` and `seed`.
|
||||
//
|
||||
// Arguments:
|
||||
// shape: The shape of the output tensor.
|
||||
// seed: 2 seeds (shape [2]).
|
||||
//
|
||||
// Returns Random values with specified shape.
|
||||
func StatelessRandomUniform(scope *Scope, shape tf.Output, seed tf.Output, optional ...StatelessRandomUniformAttr) (output tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "StatelessRandomUniform",
|
||||
Input: []tf.Input{
|
||||
shape, seed,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// An Op to exchange data across TPU replicas.
|
||||
//
|
||||
// On each replica, the input is split into `split_count` blocks along
|
||||
@ -34961,106 +35061,6 @@ func MutableHashTableOfTensorsV2(scope *Scope, key_dtype tf.DataType, value_dtyp
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2Attr is an optional argument to MutableDenseHashTableV2.
|
||||
type MutableDenseHashTableV2Attr func(optionalAttr)
|
||||
|
||||
// MutableDenseHashTableV2Container sets the optional container attribute to value.
|
||||
//
|
||||
// value: If non-empty, this table is placed in the given container.
|
||||
// Otherwise, a default container is used.
|
||||
// If not specified, defaults to ""
|
||||
func MutableDenseHashTableV2Container(value string) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["container"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2SharedName sets the optional shared_name attribute to value.
|
||||
//
|
||||
// value: If non-empty, this table is shared under the given name across
|
||||
// multiple sessions.
|
||||
// If not specified, defaults to ""
|
||||
func MutableDenseHashTableV2SharedName(value string) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["shared_name"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2UseNodeNameSharing sets the optional use_node_name_sharing attribute to value.
|
||||
// If not specified, defaults to false
|
||||
func MutableDenseHashTableV2UseNodeNameSharing(value bool) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["use_node_name_sharing"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2ValueShape sets the optional value_shape attribute to value.
|
||||
//
|
||||
// value: The shape of each value.
|
||||
// If not specified, defaults to <>
|
||||
func MutableDenseHashTableV2ValueShape(value tf.Shape) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["value_shape"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2InitialNumBuckets sets the optional initial_num_buckets attribute to value.
|
||||
//
|
||||
// value: The initial number of hash table buckets. Must be a power
|
||||
// to 2.
|
||||
// If not specified, defaults to 131072
|
||||
func MutableDenseHashTableV2InitialNumBuckets(value int64) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["initial_num_buckets"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// MutableDenseHashTableV2MaxLoadFactor sets the optional max_load_factor attribute to value.
|
||||
//
|
||||
// value: The maximum ratio between number of entries and number of
|
||||
// buckets before growing the table. Must be between 0 and 1.
|
||||
// If not specified, defaults to 0.8
|
||||
func MutableDenseHashTableV2MaxLoadFactor(value float32) MutableDenseHashTableV2Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["max_load_factor"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Creates an empty hash table that uses tensors as the backing store.
|
||||
//
|
||||
// It uses "open addressing" with quadratic reprobing to resolve
|
||||
// collisions.
|
||||
//
|
||||
// This op creates a mutable hash table, specifying the type of its keys and
|
||||
// values. Each value must be a scalar. Data can be inserted into the table using
|
||||
// the insert operations. It does not support the initialization operation.
|
||||
//
|
||||
// Arguments:
|
||||
// empty_key: The key used to represent empty key buckets internally. Must not
|
||||
// be used in insert or lookup operations.
|
||||
//
|
||||
// value_dtype: Type of the table values.
|
||||
//
|
||||
// Returns Handle to a table.
|
||||
func MutableDenseHashTableV2(scope *Scope, empty_key tf.Output, deleted_key tf.Output, value_dtype tf.DataType, optional ...MutableDenseHashTableV2Attr) (table_handle tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"value_dtype": value_dtype}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "MutableDenseHashTableV2",
|
||||
Input: []tf.Input{
|
||||
empty_key, deleted_key,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// Table initializer that takes two tensors for keys and values respectively.
|
||||
//
|
||||
// Arguments:
|
||||
|
||||
Loading…
Reference in New Issue
Block a user