Creates a dataset that reads data from the tf.data service.
Nested Classes
class | DataServiceDataset.Options | Optional attributes for DataServiceDataset
|
Public Methods
Output<Object> |
asOutput()
Returns the symbolic handle of a tensor.
|
static DataServiceDataset |
create(Scope scope, Operand<Long> datasetId, Operand<String> processingMode, Operand<String> address, Operand<String> protocol, Operand<String> jobName, Operand<Long> maxOutstandingRequests, Operand<?> iterationCounter, List<Class<?>> outputTypes, List<Shape> outputShapes, Options... options)
Factory method to create a class wrapping a new DataServiceDataset operation.
|
static DataServiceDataset.Options |
crossTrainerCacheOptions(String crossTrainerCacheOptions)
|
static DataServiceDataset.Options |
dataTransferProtocol(String dataTransferProtocol)
|
Output<?> |
handle()
|
static DataServiceDataset.Options |
targetWorkers(String targetWorkers)
|
static DataServiceDataset.Options |
taskRefreshIntervalHintMs(Long taskRefreshIntervalHintMs)
|
Inherited Methods
Public Methods
public Output<Object> asOutput ()
Returns the symbolic handle of a tensor.
Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.
public static DataServiceDataset create (Scope scope, Operand<Long> datasetId, Operand<String> processingMode, Operand<String> address, Operand<String> protocol, Operand<String> jobName, Operand<Long> maxOutstandingRequests, Operand<?> iterationCounter, List<Class<?>> outputTypes, List<Shape> outputShapes, Options... options)
Factory method to create a class wrapping a new DataServiceDataset operation.
Parameters
scope | current scope |
---|---|
options | carries optional attributes values |
Returns
- a new instance of DataServiceDataset