Go: Update generated wrapper functions for TensorFlow ops.
PiperOrigin-RevId: 297734339 Change-Id: I0e8e7cc17408ca4e59570fd27a519ac5d1383453
This commit is contained in:
A. Unique TensorFlower
TensorFlower Gardener
parent
a849a0bbac
commit
cee4c00759
@ -11472,6 +11472,87 @@ func ShardDataset(scope *Scope, input_dataset tf.Output, num_shards tf.Output, i
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// NonMaxSuppressionV5Attr is an optional argument to NonMaxSuppressionV5.
|
||||
type NonMaxSuppressionV5Attr func(optionalAttr)
|
||||
|
||||
// NonMaxSuppressionV5PadToMaxOutputSize sets the optional pad_to_max_output_size attribute to value.
|
||||
//
|
||||
// value: If true, the output `selected_indices` is padded to be of length
|
||||
// `max_output_size`. Defaults to false.
|
||||
// If not specified, defaults to false
|
||||
func NonMaxSuppressionV5PadToMaxOutputSize(value bool) NonMaxSuppressionV5Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["pad_to_max_output_size"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Greedily selects a subset of bounding boxes in descending order of score,
|
||||
//
|
||||
// pruning away boxes that have high intersection-over-union (IOU) overlap
|
||||
// with previously selected boxes. Bounding boxes with score less than
|
||||
// `score_threshold` are removed. Bounding boxes are supplied as
|
||||
// [y1, x1, y2, x2], where (y1, x1) and (y2, x2) are the coordinates of any
|
||||
// diagonal pair of box corners and the coordinates can be provided as normalized
|
||||
// (i.e., lying in the interval [0, 1]) or absolute. Note that this algorithm
|
||||
// is agnostic to where the origin is in the coordinate system and more
|
||||
// generally is invariant to orthogonal transformations and translations
|
||||
// of the coordinate system; thus translating or reflections of the coordinate
|
||||
// system result in the same boxes being selected by the algorithm.
|
||||
// The output of this operation is a set of integers indexing into the input
|
||||
// collection of bounding boxes representing the selected boxes. The bounding
|
||||
// box coordinates corresponding to the selected indices can then be obtained
|
||||
// using the `tf.gather operation`. For example:
|
||||
// selected_indices = tf.image.non_max_suppression_v2(
|
||||
// boxes, scores, max_output_size, iou_threshold, score_threshold)
|
||||
// selected_boxes = tf.gather(boxes, selected_indices)
|
||||
// This op also supports a Soft-NMS (with Gaussian weighting) mode (c.f.
|
||||
// Bodla et al, https://arxiv.org/abs/1704.04503) where boxes reduce the score
|
||||
// of other overlapping boxes instead of directly causing them to be pruned.
|
||||
// To enable this Soft-NMS mode, set the `soft_nms_sigma` parameter to be
|
||||
// larger than 0.
|
||||
//
|
||||
// Arguments:
|
||||
// boxes: A 2-D float tensor of shape `[num_boxes, 4]`.
|
||||
// scores: A 1-D float tensor of shape `[num_boxes]` representing a single
|
||||
// score corresponding to each box (each row of boxes).
|
||||
// max_output_size: A scalar integer tensor representing the maximum number of
|
||||
// boxes to be selected by non max suppression.
|
||||
// iou_threshold: A 0-D float tensor representing the threshold for deciding whether
|
||||
// boxes overlap too much with respect to IOU.
|
||||
// score_threshold: A 0-D float tensor representing the threshold for deciding when to remove
|
||||
// boxes based on score.
|
||||
// soft_nms_sigma: A 0-D float tensor representing the sigma parameter for Soft NMS; see Bodla et
|
||||
// al (c.f. https://arxiv.org/abs/1704.04503). When `soft_nms_sigma=0.0` (which
|
||||
// is default), we fall back to standard (hard) NMS.
|
||||
//
|
||||
// Returns:
|
||||
// selected_indices: A 1-D integer tensor of shape `[M]` representing the selected
|
||||
// indices from the boxes tensor, where `M <= max_output_size`.
|
||||
// selected_scores: A 1-D float tensor of shape `[M]` representing the corresponding
|
||||
// scores for each selected box, where `M <= max_output_size`. Scores only differ
|
||||
// from corresponding input scores when using Soft NMS (i.e. when
|
||||
// `soft_nms_sigma>0`)
|
||||
// valid_outputs: A 0-D integer tensor representing the number of valid elements in
|
||||
// `selected_indices`, with the valid elements appearing first.
|
||||
func NonMaxSuppressionV5(scope *Scope, boxes tf.Output, scores tf.Output, max_output_size tf.Output, iou_threshold tf.Output, score_threshold tf.Output, soft_nms_sigma tf.Output, optional ...NonMaxSuppressionV5Attr) (selected_indices tf.Output, selected_scores tf.Output, valid_outputs tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "NonMaxSuppressionV5",
|
||||
Input: []tf.Input{
|
||||
boxes, scores, max_output_size, iou_threshold, score_threshold, soft_nms_sigma,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0), op.Output(1), op.Output(2)
|
||||
}
|
||||
|
||||
// NonMaxSuppressionV4Attr is an optional argument to NonMaxSuppressionV4.
|
||||
type NonMaxSuppressionV4Attr func(optionalAttr)
|
||||
|
||||
@ -19262,75 +19343,6 @@ func Greater(scope *Scope, x tf.Output, y tf.Output) (z tf.Output) {
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// A container for a multi device iterator resource.
|
||||
//
|
||||
// Returns:
|
||||
// handle: A handle to a multi device iterator that can be passed to a
|
||||
// "MultiDeviceIteratorGetNextFromShard" op. In contrast to MultiDeviceIterator,
|
||||
// AnonymousIterator prevents resource sharing by name, and does not keep a
|
||||
// reference to the resource container.
|
||||
// deleter: A variant deleter that should be passed into the op that deletes the iterator.
|
||||
func AnonymousMultiDeviceIterator(scope *Scope, devices []string, output_types []tf.DataType, output_shapes []tf.Shape) (handle tf.Output, deleter tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"devices": devices, "output_types": output_types, "output_shapes": output_shapes}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "AnonymousMultiDeviceIterator",
|
||||
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0), op.Output(1)
|
||||
}
|
||||
|
||||
// Provides the time since epoch in seconds.
|
||||
//
|
||||
// Returns the timestamp as a `float64` for seconds since the Unix epoch.
|
||||
//
|
||||
// Note: the timestamp is computed when the op is executed, not when it is added
|
||||
// to the graph.
|
||||
func Timestamp(scope *Scope) (ts tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "Timestamp",
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// Returns the truth value of (x <= y) element-wise.
|
||||
//
|
||||
// *NOTE*: `LessEqual` supports broadcasting. More about broadcasting
|
||||
// [here](http://docs.scipy.org/doc/numpy/user/basics.broadcasting.html)
|
||||
//
|
||||
// Example:
|
||||
//
|
||||
// ```python
|
||||
// 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]
|
||||
// ```
|
||||
func LessEqual(scope *Scope, x tf.Output, y tf.Output) (z tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "LessEqual",
|
||||
Input: []tf.Input{
|
||||
x, y,
|
||||
},
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// Compute the polygamma function \\(\psi^{(n)}(x)\\).
|
||||
//
|
||||
// The polygamma function is defined as:
|
||||
@ -22833,6 +22845,196 @@ func MatrixDiagPartV2(scope *Scope, input tf.Output, k tf.Output, padding_value
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// A container for a multi device iterator resource.
|
||||
//
|
||||
// Returns:
|
||||
// handle: A handle to a multi device iterator that can be passed to a
|
||||
// "MultiDeviceIteratorGetNextFromShard" op. In contrast to MultiDeviceIterator,
|
||||
// AnonymousIterator prevents resource sharing by name, and does not keep a
|
||||
// reference to the resource container.
|
||||
// deleter: A variant deleter that should be passed into the op that deletes the iterator.
|
||||
func AnonymousMultiDeviceIterator(scope *Scope, devices []string, output_types []tf.DataType, output_shapes []tf.Shape) (handle tf.Output, deleter tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"devices": devices, "output_types": output_types, "output_shapes": output_shapes}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "AnonymousMultiDeviceIterator",
|
||||
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0), op.Output(1)
|
||||
}
|
||||
|
||||
// Provides the time since epoch in seconds.
|
||||
//
|
||||
// Returns the timestamp as a `float64` for seconds since the Unix epoch.
|
||||
//
|
||||
// Note: the timestamp is computed when the op is executed, not when it is added
|
||||
// to the graph.
|
||||
func Timestamp(scope *Scope) (ts tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "Timestamp",
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// Returns the truth value of (x <= y) element-wise.
|
||||
//
|
||||
// *NOTE*: `LessEqual` supports broadcasting. More about broadcasting
|
||||
// [here](http://docs.scipy.org/doc/numpy/user/basics.broadcasting.html)
|
||||
//
|
||||
// Example:
|
||||
//
|
||||
// ```python
|
||||
// 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]
|
||||
// ```
|
||||
func LessEqual(scope *Scope, x tf.Output, y tf.Output) (z tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "LessEqual",
|
||||
Input: []tf.Input{
|
||||
x, y,
|
||||
},
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// LoadTPUEmbeddingADAMParametersGradAccumDebugAttr is an optional argument to LoadTPUEmbeddingADAMParametersGradAccumDebug.
|
||||
type LoadTPUEmbeddingADAMParametersGradAccumDebugAttr func(optionalAttr)
|
||||
|
||||
// LoadTPUEmbeddingADAMParametersGradAccumDebugTableId sets the optional table_id attribute to value.
|
||||
// If not specified, defaults to -1
|
||||
//
|
||||
// REQUIRES: value >= -1
|
||||
func LoadTPUEmbeddingADAMParametersGradAccumDebugTableId(value int64) LoadTPUEmbeddingADAMParametersGradAccumDebugAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["table_id"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// LoadTPUEmbeddingADAMParametersGradAccumDebugTableName sets the optional table_name attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func LoadTPUEmbeddingADAMParametersGradAccumDebugTableName(value string) LoadTPUEmbeddingADAMParametersGradAccumDebugAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["table_name"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// LoadTPUEmbeddingADAMParametersGradAccumDebugConfig sets the optional config attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func LoadTPUEmbeddingADAMParametersGradAccumDebugConfig(value string) LoadTPUEmbeddingADAMParametersGradAccumDebugAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["config"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Load ADAM embedding parameters with debug support.
|
||||
//
|
||||
// An op that loads optimization parameters into HBM for embedding. Must be
|
||||
// preceded by a ConfigureTPUEmbeddingHost op that sets up the correct
|
||||
// embedding table configuration. For example, this op is used to install
|
||||
// parameters that are loaded from a checkpoint before a training loop is
|
||||
// executed.
|
||||
//
|
||||
// Arguments:
|
||||
// parameters: Value of parameters used in the ADAM optimization algorithm.
|
||||
// momenta: Value of momenta used in the ADAM optimization algorithm.
|
||||
// velocities: Value of velocities used in the ADAM optimization algorithm.
|
||||
// gradient_accumulators: Value of gradient_accumulators used in the ADAM optimization algorithm.
|
||||
//
|
||||
//
|
||||
//
|
||||
// Returns the created operation.
|
||||
func LoadTPUEmbeddingADAMParametersGradAccumDebug(scope *Scope, parameters tf.Output, momenta tf.Output, velocities tf.Output, gradient_accumulators tf.Output, num_shards int64, shard_id int64, optional ...LoadTPUEmbeddingADAMParametersGradAccumDebugAttr) (o *tf.Operation) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"num_shards": num_shards, "shard_id": shard_id}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "LoadTPUEmbeddingADAMParametersGradAccumDebug",
|
||||
Input: []tf.Input{
|
||||
parameters, momenta, velocities, gradient_accumulators,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
return scope.AddOperation(opspec)
|
||||
}
|
||||
|
||||
// RetrieveTPUEmbeddingRMSPropParametersAttr is an optional argument to RetrieveTPUEmbeddingRMSPropParameters.
|
||||
type RetrieveTPUEmbeddingRMSPropParametersAttr func(optionalAttr)
|
||||
|
||||
// RetrieveTPUEmbeddingRMSPropParametersTableId sets the optional table_id attribute to value.
|
||||
// If not specified, defaults to -1
|
||||
//
|
||||
// REQUIRES: value >= -1
|
||||
func RetrieveTPUEmbeddingRMSPropParametersTableId(value int64) RetrieveTPUEmbeddingRMSPropParametersAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["table_id"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// RetrieveTPUEmbeddingRMSPropParametersTableName sets the optional table_name attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func RetrieveTPUEmbeddingRMSPropParametersTableName(value string) RetrieveTPUEmbeddingRMSPropParametersAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["table_name"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// RetrieveTPUEmbeddingRMSPropParametersConfig sets the optional config attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func RetrieveTPUEmbeddingRMSPropParametersConfig(value string) RetrieveTPUEmbeddingRMSPropParametersAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["config"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Retrieve RMSProp embedding parameters.
|
||||
//
|
||||
// An op that retrieves optimization parameters from embedding to host
|
||||
// memory. Must be preceded by a ConfigureTPUEmbeddingHost op that sets up
|
||||
// the correct embedding table configuration. For example, this op is
|
||||
// used to retrieve updated parameters before saving a checkpoint.
|
||||
//
|
||||
// Returns:
|
||||
// parameters: Parameter parameters updated by the RMSProp optimization algorithm.
|
||||
// ms: Parameter ms updated by the RMSProp optimization algorithm.
|
||||
// mom: Parameter mom updated by the RMSProp optimization algorithm.
|
||||
func RetrieveTPUEmbeddingRMSPropParameters(scope *Scope, num_shards int64, shard_id int64, optional ...RetrieveTPUEmbeddingRMSPropParametersAttr) (parameters tf.Output, ms tf.Output, mom tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"num_shards": num_shards, "shard_id": shard_id}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "RetrieveTPUEmbeddingRMSPropParameters",
|
||||
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0), op.Output(1), op.Output(2)
|
||||
}
|
||||
|
||||
// Conv3DBackpropInputV2Attr is an optional argument to Conv3DBackpropInputV2.
|
||||
type Conv3DBackpropInputV2Attr func(optionalAttr)
|
||||
|
||||
@ -41212,127 +41414,6 @@ func BesselI1e(scope *Scope, x tf.Output) (y tf.Output) {
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// LoadTPUEmbeddingADAMParametersGradAccumDebugAttr is an optional argument to LoadTPUEmbeddingADAMParametersGradAccumDebug.
|
||||
type LoadTPUEmbeddingADAMParametersGradAccumDebugAttr func(optionalAttr)
|
||||
|
||||
// LoadTPUEmbeddingADAMParametersGradAccumDebugTableId sets the optional table_id attribute to value.
|
||||
// If not specified, defaults to -1
|
||||
//
|
||||
// REQUIRES: value >= -1
|
||||
func LoadTPUEmbeddingADAMParametersGradAccumDebugTableId(value int64) LoadTPUEmbeddingADAMParametersGradAccumDebugAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["table_id"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// LoadTPUEmbeddingADAMParametersGradAccumDebugTableName sets the optional table_name attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func LoadTPUEmbeddingADAMParametersGradAccumDebugTableName(value string) LoadTPUEmbeddingADAMParametersGradAccumDebugAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["table_name"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// LoadTPUEmbeddingADAMParametersGradAccumDebugConfig sets the optional config attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func LoadTPUEmbeddingADAMParametersGradAccumDebugConfig(value string) LoadTPUEmbeddingADAMParametersGradAccumDebugAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["config"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Load ADAM embedding parameters with debug support.
|
||||
//
|
||||
// An op that loads optimization parameters into HBM for embedding. Must be
|
||||
// preceded by a ConfigureTPUEmbeddingHost op that sets up the correct
|
||||
// embedding table configuration. For example, this op is used to install
|
||||
// parameters that are loaded from a checkpoint before a training loop is
|
||||
// executed.
|
||||
//
|
||||
// Arguments:
|
||||
// parameters: Value of parameters used in the ADAM optimization algorithm.
|
||||
// momenta: Value of momenta used in the ADAM optimization algorithm.
|
||||
// velocities: Value of velocities used in the ADAM optimization algorithm.
|
||||
// gradient_accumulators: Value of gradient_accumulators used in the ADAM optimization algorithm.
|
||||
//
|
||||
//
|
||||
//
|
||||
// Returns the created operation.
|
||||
func LoadTPUEmbeddingADAMParametersGradAccumDebug(scope *Scope, parameters tf.Output, momenta tf.Output, velocities tf.Output, gradient_accumulators tf.Output, num_shards int64, shard_id int64, optional ...LoadTPUEmbeddingADAMParametersGradAccumDebugAttr) (o *tf.Operation) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"num_shards": num_shards, "shard_id": shard_id}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "LoadTPUEmbeddingADAMParametersGradAccumDebug",
|
||||
Input: []tf.Input{
|
||||
parameters, momenta, velocities, gradient_accumulators,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
return scope.AddOperation(opspec)
|
||||
}
|
||||
|
||||
// RetrieveTPUEmbeddingRMSPropParametersAttr is an optional argument to RetrieveTPUEmbeddingRMSPropParameters.
|
||||
type RetrieveTPUEmbeddingRMSPropParametersAttr func(optionalAttr)
|
||||
|
||||
// RetrieveTPUEmbeddingRMSPropParametersTableId sets the optional table_id attribute to value.
|
||||
// If not specified, defaults to -1
|
||||
//
|
||||
// REQUIRES: value >= -1
|
||||
func RetrieveTPUEmbeddingRMSPropParametersTableId(value int64) RetrieveTPUEmbeddingRMSPropParametersAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["table_id"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// RetrieveTPUEmbeddingRMSPropParametersTableName sets the optional table_name attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func RetrieveTPUEmbeddingRMSPropParametersTableName(value string) RetrieveTPUEmbeddingRMSPropParametersAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["table_name"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// RetrieveTPUEmbeddingRMSPropParametersConfig sets the optional config attribute to value.
|
||||
// If not specified, defaults to ""
|
||||
func RetrieveTPUEmbeddingRMSPropParametersConfig(value string) RetrieveTPUEmbeddingRMSPropParametersAttr {
|
||||
return func(m optionalAttr) {
|
||||
m["config"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Retrieve RMSProp embedding parameters.
|
||||
//
|
||||
// An op that retrieves optimization parameters from embedding to host
|
||||
// memory. Must be preceded by a ConfigureTPUEmbeddingHost op that sets up
|
||||
// the correct embedding table configuration. For example, this op is
|
||||
// used to retrieve updated parameters before saving a checkpoint.
|
||||
//
|
||||
// Returns:
|
||||
// parameters: Parameter parameters updated by the RMSProp optimization algorithm.
|
||||
// ms: Parameter ms updated by the RMSProp optimization algorithm.
|
||||
// mom: Parameter mom updated by the RMSProp optimization algorithm.
|
||||
func RetrieveTPUEmbeddingRMSPropParameters(scope *Scope, num_shards int64, shard_id int64, optional ...RetrieveTPUEmbeddingRMSPropParametersAttr) (parameters tf.Output, ms tf.Output, mom tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{"num_shards": num_shards, "shard_id": shard_id}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "RetrieveTPUEmbeddingRMSPropParameters",
|
||||
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0), op.Output(1), op.Output(2)
|
||||
}
|
||||
|
||||
// Returns a batched diagonal tensor with a given batched diagonal values.
|
||||
//
|
||||
// Given a `diagonal`, this operation returns a tensor with the `diagonal` and
|
||||
@ -42267,87 +42348,6 @@ func SerializeManySparse(scope *Scope, sparse_indices tf.Output, sparse_values t
|
||||
return op.Output(0)
|
||||
}
|
||||
|
||||
// NonMaxSuppressionV5Attr is an optional argument to NonMaxSuppressionV5.
|
||||
type NonMaxSuppressionV5Attr func(optionalAttr)
|
||||
|
||||
// NonMaxSuppressionV5PadToMaxOutputSize sets the optional pad_to_max_output_size attribute to value.
|
||||
//
|
||||
// value: If true, the output `selected_indices` is padded to be of length
|
||||
// `max_output_size`. Defaults to false.
|
||||
// If not specified, defaults to false
|
||||
func NonMaxSuppressionV5PadToMaxOutputSize(value bool) NonMaxSuppressionV5Attr {
|
||||
return func(m optionalAttr) {
|
||||
m["pad_to_max_output_size"] = value
|
||||
}
|
||||
}
|
||||
|
||||
// Greedily selects a subset of bounding boxes in descending order of score,
|
||||
//
|
||||
// pruning away boxes that have high intersection-over-union (IOU) overlap
|
||||
// with previously selected boxes. Bounding boxes with score less than
|
||||
// `score_threshold` are removed. Bounding boxes are supplied as
|
||||
// [y1, x1, y2, x2], where (y1, x1) and (y2, x2) are the coordinates of any
|
||||
// diagonal pair of box corners and the coordinates can be provided as normalized
|
||||
// (i.e., lying in the interval [0, 1]) or absolute. Note that this algorithm
|
||||
// is agnostic to where the origin is in the coordinate system and more
|
||||
// generally is invariant to orthogonal transformations and translations
|
||||
// of the coordinate system; thus translating or reflections of the coordinate
|
||||
// system result in the same boxes being selected by the algorithm.
|
||||
// The output of this operation is a set of integers indexing into the input
|
||||
// collection of bounding boxes representing the selected boxes. The bounding
|
||||
// box coordinates corresponding to the selected indices can then be obtained
|
||||
// using the `tf.gather operation`. For example:
|
||||
// selected_indices = tf.image.non_max_suppression_v2(
|
||||
// boxes, scores, max_output_size, iou_threshold, score_threshold)
|
||||
// selected_boxes = tf.gather(boxes, selected_indices)
|
||||
// This op also supports a Soft-NMS (with Gaussian weighting) mode (c.f.
|
||||
// Bodla et al, https://arxiv.org/abs/1704.04503) where boxes reduce the score
|
||||
// of other overlapping boxes instead of directly causing them to be pruned.
|
||||
// To enable this Soft-NMS mode, set the `soft_nms_sigma` parameter to be
|
||||
// larger than 0.
|
||||
//
|
||||
// Arguments:
|
||||
// boxes: A 2-D float tensor of shape `[num_boxes, 4]`.
|
||||
// scores: A 1-D float tensor of shape `[num_boxes]` representing a single
|
||||
// score corresponding to each box (each row of boxes).
|
||||
// max_output_size: A scalar integer tensor representing the maximum number of
|
||||
// boxes to be selected by non max suppression.
|
||||
// iou_threshold: A 0-D float tensor representing the threshold for deciding whether
|
||||
// boxes overlap too much with respect to IOU.
|
||||
// score_threshold: A 0-D float tensor representing the threshold for deciding when to remove
|
||||
// boxes based on score.
|
||||
// soft_nms_sigma: A 0-D float tensor representing the sigma parameter for Soft NMS; see Bodla et
|
||||
// al (c.f. https://arxiv.org/abs/1704.04503). When `soft_nms_sigma=0.0` (which
|
||||
// is default), we fall back to standard (hard) NMS.
|
||||
//
|
||||
// Returns:
|
||||
// selected_indices: A 1-D integer tensor of shape `[M]` representing the selected
|
||||
// indices from the boxes tensor, where `M <= max_output_size`.
|
||||
// selected_scores: A 1-D float tensor of shape `[M]` representing the corresponding
|
||||
// scores for each selected box, where `M <= max_output_size`. Scores only differ
|
||||
// from corresponding input scores when using Soft NMS (i.e. when
|
||||
// `soft_nms_sigma>0`)
|
||||
// valid_outputs: A 0-D integer tensor representing the number of valid elements in
|
||||
// `selected_indices`, with the valid elements appearing first.
|
||||
func NonMaxSuppressionV5(scope *Scope, boxes tf.Output, scores tf.Output, max_output_size tf.Output, iou_threshold tf.Output, score_threshold tf.Output, soft_nms_sigma tf.Output, optional ...NonMaxSuppressionV5Attr) (selected_indices tf.Output, selected_scores tf.Output, valid_outputs tf.Output) {
|
||||
if scope.Err() != nil {
|
||||
return
|
||||
}
|
||||
attrs := map[string]interface{}{}
|
||||
for _, a := range optional {
|
||||
a(attrs)
|
||||
}
|
||||
opspec := tf.OpSpec{
|
||||
Type: "NonMaxSuppressionV5",
|
||||
Input: []tf.Input{
|
||||
boxes, scores, max_output_size, iou_threshold, score_threshold, soft_nms_sigma,
|
||||
},
|
||||
Attrs: attrs,
|
||||
}
|
||||
op := scope.AddOperation(opspec)
|
||||
return op.Output(0), op.Output(1), op.Output(2)
|
||||
}
|
||||
|
||||
// Says whether the targets are in the top `K` predictions.
|
||||
//
|
||||
// This outputs a `batch_size` bool array, an entry `out[i]` is `true` if the
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user