data

Type Members

case class BatchDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], batchSize: Long, name: String = "BatchDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the batch op.
Dataset that wraps the application of the batch op.
$OpDocDatasetBatch
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
batchSize
Batch size to use.
name
Name for this dataset.
case class CacheDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], directory: String, name: String = "CacheDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the cache op.
Dataset that wraps the application of the cache op.
$OpDocDatasetCache
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
directory
Directory to use for caching. If empty, then the provided dataset will be cached in memory.
name
Name for this dataset.
case class ConcatenatedDataset[T, O, D, S](inputDataset1: Dataset[T, O, D, S], inputDataset2: Dataset[T, O, D, S], name: String = "ConcatenatedDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the concatenate op.
Dataset that wraps the application of the concatenate op.
$OpDocDatasetConcatenate
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset1
First input dataset.
inputDataset2
Second input dataset.
name
Name for this dataset.

Annotations
@throws( ... )
Exceptions thrown
IllegalArgumentException If the data types of the input datasets are not identical of if their shapes are not compatible.
trait Data[T] extends AnyRef

Data can be emitted by Datasets (i.e., the element types of all Datasets are Data).
Data can be emitted by Datasets (i.e., the element types of all Datasets are Data).
Currently supported data types are:
- Single Tensor.
- Sequences of other Data (e.g., Seqs, Lists, etc.).
  - Sequences that are not homogeneous are not supported (e.g., Seq(data1, Seq(data1, data2))).
  - Note that, for that reason, even though Seq(List(data1), List(data1, data2)) is supported, Seq(Seq(data1), List(data1, data2)) is not.
  - A sequence containing both Outputs and SparseOutputs, for example, is considered heterogeneous. For such cases, it is advisable to use tuples.
- Arrays of other Data.
- Maps with arbitrary key types and Data value types.
- Products of other Data (e.g., tuples).
  - Note that with tuples, heterogeneous types are supported, due to the tuple type being a kind of heterogeneous collection. Internally, the data emitted by a Dataset will be de-duplicated to prevent redundant computation.
This trait guarantees that the output data types and shapes of a Dataset will match the structure of the corresponding data. For example, if a Seq(List(data1), List(data1, data2)) is provided as a Dataset element type, then the dataset output data types will have the following structure Seq(List(type1), List(type1, type2)), and similarly for the output shapes.
abstract class Dataset[T, O, D, S] extends AnyRef

Represents a potentially large set of elements.
Represents a potentially large set of elements.
A dataset can be used to represent an input pipeline as a collection of elements (i.e., nested structures of tensors) and a "logical plan" of transformations that act on those elements.
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
case class DropDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Long, name: String = "DropDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the drop op.
Dataset that wraps the application of the drop op.
$OpDocDatasetDrop
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
count
Number of elements to drop.
name
Name for this dataset.
case class DynamicDropDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Output, name: String = "DropDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the drop op.
Dataset that wraps the application of the drop op.
$OpDocDatasetDrop
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
count
Number of elements to drop.
name
Name for this dataset.
case class DynamicPaddedBatchDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], batchSize: Long, paddedShapes: S, paddingValues: O = null.asInstanceOf[O], name: String = "PaddedBatchDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the paddedBatch op.
Dataset that wraps the application of the paddedBatch op.
$OpDocDatasetPaddedBatch
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
batchSize
Batch size to use.
paddedShapes
Shape to which the respective component of each input element should be padded prior to batching. Any unknown dimensions (e.g., equal to -1) will be padded to the maximum size of that dimension in each batch.
paddingValues
Scalar tensor structure representing the padding values to use for the respective components. Defaults to zero for numeric types and the empty string for string types.
name
Name for this dataset.
class DynamicTFRecordDataset extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape]

Dataset with elements read from TensorFlow record files.
case class DynamicTakeDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Output, name: String = "TakeDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the take op.
Dataset that wraps the application of the take op.
$OpDocDatasetTake
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
count
Number of elements to take.
name
Name for this dataset.
case class DynamicTextLinesDataset(filenames: Output, compressionType: CompressionType = NoCompression, bufferSize: Long = 256 * 1024, name: String = "TextLineDataset") extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape] with Product with Serializable

Dataset with elements read from text files (each line in each file corresponds to an element).
Dataset with elements read from text files (each line in each file corresponds to an element).
**Note:** New-line characters are stripped from the output.
filenames
Scalar or vector tensor containing the the name(s) of the file(s) to be read.
compressionType
Compression type for the file.
bufferSize
Number of bytes to buffer while reading from the file.
name
Name for this dataset.
case class FilterDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], predicateFn: (O) ⇒ Output, name: String = "FilterDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the filter op.
Dataset that wraps the application of the filter op.
$OpDocDatasetFilter
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
predicateFn
Filter predicate function.
name
Name for this dataset.
case class FixedLengthRecordDataset(filenames: tensors.Tensor[types.STRING], recordNumBytes: Long, headerNumBytes: Long, footerNumBytes: Long, bufferSize: Long = 256 * 1024, name: String = "FixedLengthRecordDataset") extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape] with Product with Serializable

Dataset with elements read from binary files.
Dataset with elements read from binary files.
filenames
Scalar or vector tensor containing the the name(s) of the file(s) to be read.
recordNumBytes
Number of bytes in the record.
headerNumBytes
Number of bytes in the header (i.e., the number of bytes to skip at the beginning of a file).
footerNumBytes
Number of bytes in the footer (i.e., the number of bytes to skip at the end of a file).
bufferSize
Number of bytes to buffer while reading from the file.
name
Name for this dataset.
case class FlatMapDataset[T, O, D, S, RT, RO, RD, RS](inputDataset: Dataset[T, O, D, S], function: (O) ⇒ Dataset[RT, RO, RD, RS], name: String = "FlatMapDataset")(implicit evData: Aux[T, O, D, S] = inputDataset.evData, evFunctionInput: ArgType[O] = inputDataset.evFunctionInput, evStructure: Aux[RT, RO, RD, RS], evRData: Aux[RT, RO, RD, RS], evFunctionOutput: ArgType[RO]) extends Dataset[RT, RO, RD, RS] with Product with Serializable

Dataset that wraps the application of the flatMap op.
Dataset that wraps the application of the flatMap op.
$OpDocDatasetFlatMap
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
function
Mapping function.
name
Name for this dataset.
case class GroupByWindowDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], keyFn: (O) ⇒ Output, reduceFn: ((Output, Dataset[T, O, D, S])) ⇒ Dataset[T, O, D, S], windowSizeFn: (Output) ⇒ Output, name: String = "GroupByWindowDataset")(implicit evStructure: Aux[T, O, D, S]) extends Dataset[T, O, D, S] with Product with Serializable

$OpDocDatasetGroupByWindow
$OpDocDatasetGroupByWindow
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
keyFn
Function used to compute the grouping key.
reduceFn
Function used to reduce each group.
windowSizeFn
Function used to compute the maximum window size per key.
name
Name for this dataset.
case class IgnoreErrorsDataset[T, O, D, S] extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the ignoreErrors op.
Dataset that wraps the application of the ignoreErrors op.
$OpDocDatasetIgnoreErrors
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
class InitializableIterator[T, O, D, S] extends Iterator[T, O, D, S]

An iterator that contains an initializer.
An iterator that contains an initializer.
An iterator represents the state of iterating through a dataset.
case class InterleaveDataset[T, O, D, S, RT, RO, RD, RS](inputDataset: Dataset[T, O, D, S], function: (O) ⇒ Dataset[RT, RO, RD, RS], cycleLength: Output, blockLength: Output = 1, name: String = "InterleaveDataset")(implicit evData: Aux[T, O, D, S] = inputDataset.evData, evFunctionInput: ArgType[O] = inputDataset.evFunctionInput, evStructure: Aux[RT, RO, RD, RS], evRData: Aux[RT, RO, RD, RS], evFunctionOutput: ArgType[RO]) extends Dataset[RT, RO, RD, RS] with Product with Serializable

Dataset that wraps the application of the interleave op.
Dataset that wraps the application of the interleave op.
$OpDocDatasetInterleave
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
function
Mapping function.
cycleLength
Number of elements from the input dataset that will be processed concurrently.
blockLength
Number of consecutive elements to produce from each input element before cycling to another input element.
name
Name for this dataset.
class Iterator[T, O, D, S] extends AnyRef

A simple iterator that does contains an initializer and can thus not be used until an initializer is created for it, using its createInitializer method.
A simple iterator that does contains an initializer and can thus not be used until an initializer is created for it, using its createInitializer method.
An iterator represents the state of iterating through a dataset.
case class MapDataset[T, O, D, S, RT, RO, RD, RS](inputDataset: Dataset[T, O, D, S], function: (O) ⇒ RO, name: String = "MapDataset")(implicit evData: Aux[T, O, D, S] = inputDataset.evData, evFunctionInput: ArgType[O] = inputDataset.evFunctionInput, evStructure: Aux[RT, RO, RD, RS], evRData: Aux[RT, RO, RD, RS], evFunctionOutput: ArgType[RO]) extends Dataset[RT, RO, RD, RS] with Product with Serializable

Dataset that wraps the application of the map op.
Dataset that wraps the application of the map op.
$OpDocDatasetMap
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
function
Mapping function.
name
Name for this dataset.
case class OutputDataset[T, O, D, S](data: O, name: String = "OutputDataset")(implicit evStructure: Aux[T, O, D, S], evData: Aux[T, O, D, S], evFunctionInput: ArgType[O]) extends Dataset[T, O, D, S] with Product with Serializable

Dataset with a single element.
Dataset with a single element.
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
data
Data representing the single element of this dataset.
name
Name for this dataset.
case class OutputSlicesDataset[T, O, D, S](data: O, name: String = "OutputSlicesDataset")(implicit evStructure: Aux[T, O, D, S], evData: Aux[T, O, D, S], evFunctionInput: ArgType[O]) extends Dataset[T, O, D, S] with Product with Serializable

Dataset with slices from the nested structure of Outputs (i.e., a Data-supported type).
Dataset with slices from the nested structure of Outputs (i.e., a Data-supported type). The slices are taken along the first axis of each Output in the nested structure.
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
data
Data representing the elements of this dataset.
name
Name for this dataset.
case class PaddedBatchDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], batchSize: Long, paddedShapes: S, paddingValues: T = null.asInstanceOf[T], name: String = "PaddedBatchDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the paddedBatch op.
Dataset that wraps the application of the paddedBatch op.
$OpDocDatasetPaddedBatch
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
batchSize
Batch size to use.
paddedShapes
Shape to which the respective component of each input element should be padded prior to batching. Any unknown dimensions (e.g., equal to -1) will be padded to the maximum size of that dimension in each batch.
paddingValues
Scalar tensor structure representing the padding values to use for the respective components. Defaults to zero for numeric types and the empty string for string types.
name
Name for this dataset.
case class ParallelInterleaveDataset[T, O, D, S, RT, RO, RD, RS](inputDataset: Dataset[T, O, D, S], function: (O) ⇒ Dataset[RT, RO, RD, RS], cycleLength: Output, blockLength: Output = 1, sloppy: Boolean = false, bufferOutputElements: Output = null, prefetchInputElements: Output = null, name: String = "ParallelInterleaveDataset")(implicit evData: Aux[T, O, D, S] = inputDataset.evData, evFunctionInput: ArgType[O] = inputDataset.evFunctionInput, evStructure: Aux[RT, RO, RD, RS], evRData: Aux[RT, RO, RD, RS], evFunctionOutput: ArgType[RO]) extends Dataset[RT, RO, RD, RS] with Product with Serializable

Dataset that wraps the application of the parallelInterleave op.
Dataset that wraps the application of the parallelInterleave op.
$OpDocDatasetParallelInterleave
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
function
Mapping function.
cycleLength
Number of elements from the input dataset that will be processed concurrently.
blockLength
Number of consecutive elements to produce from each input element before cycling to another input element.
sloppy
If false, elements are produced in deterministic order. Otherwise, the implementation is allowed, for the sake of expediency, to produce elements in a non-deterministic order.
bufferOutputElements
Number of elements each iterator being interleaved should buffer (similar to the prefetch(...) transformation for each interleaved iterator).
prefetchInputElements
Number of input elements to transform to iterators before they are needed for interleaving.
name
Name for this dataset.
case class ParallelMapDataset[T, O, D, S, RT, RO, RD, RS](inputDataset: Dataset[T, O, D, S], function: (O) ⇒ RO, numParallelCalls: Int, name: String = "ParallelMapDataset")(implicit evData: Aux[T, O, D, S] = inputDataset.evData, evFunctionInput: ArgType[O] = inputDataset.evFunctionInput, evStructure: Aux[RT, RO, RD, RS], evRData: Aux[RT, RO, RD, RS], evFunctionOutput: ArgType[RO]) extends Dataset[RT, RO, RD, RS] with Product with Serializable

Dataset that wraps the application of the parallelMap op.
Dataset that wraps the application of the parallelMap op.
$OpDocDatasetMap
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
function
Mapping function.
numParallelCalls
Number of concurrent invocations of function that process elements from inputDataset in parallel.
name
Name for this dataset.
case class PrefetchDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], bufferSize: Long, name: String = "PrefetchDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the prefetch op.
Dataset that wraps the application of the prefetch op.
$OpDocDatasetPrefetch
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
bufferSize
Number of elements to prefetch.
name
Name for this dataset.
case class RangeDataset(start: Long, limit: Long, delta: Long = 1L, name: String = "RangeDataset") extends Dataset[tensors.Tensor[types.INT64], Output, types.INT64, core.Shape] with Product with Serializable

Dataset that wraps the application of the range op.
Dataset that wraps the application of the range op.
$OpDocDatasetRange
start
Starting value of the number sequence.
limit
Ending value (exclusive) of the number sequence.
delta
Difference between consecutive numbers in the sequence.
name
Name for this dataset.
case class RepeatDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Long, name: String = "RepeatDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the repeat op.
Dataset that wraps the application of the repeat op.
$OpDocDatasetRepeat
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
count
Number of times to repeat the input dataset. A value of -1 corresponds to repeating it indefinitely.
name
Name for this dataset.
case class ShuffleDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], bufferSize: Long, seed: Option[Int], name: String = "ShuffleDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the shuffle op.
Dataset that wraps the application of the shuffle op.
$OpDocDatasetShuffle
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
bufferSize
Buffer size, meaning the number of output elements to buffer in an iterator over this dataset.
seed
Seed value for the random number generator. If not provided, a random seed is used.
name
Name for this dataset.
case class SparseOutputSlicesDataset(tensor: SparseOutput, name: String = "SparseOutputSliceDataset") extends Dataset[tensors.SparseTensor[types.DataType], SparseOutput, (types.INT64, types.DataType, types.INT64), (core.Shape, core.Shape, core.Shape)] with Product with Serializable

Dataset that splits a sparse tensor into its rows.
Dataset that splits a sparse tensor into its rows.
tensor
Sparse tensor.
name
Name for this dataset.
case class SparseTensorSlicesDataset[D <: types.DataType](tensor: tensors.SparseTensor[D], name: String = "SparseTensorSliceDataset") extends Dataset[tensors.SparseTensor[D], SparseOutput, (types.INT64, D, types.INT64), (core.Shape, core.Shape, core.Shape)] with Product with Serializable

Dataset that splits a sparse tensor into its rows.
Dataset that splits a sparse tensor into its rows.
tensor
Sparse tensor.
name
Name for this dataset.
class TFRecordDataset extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape]

Dataset with elements read from TensorFlow record files.
case class TakeDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Long, name: String = "TakeDataset") extends Dataset[T, O, D, S] with Product with Serializable

Dataset that wraps the application of the take op.
Dataset that wraps the application of the take op.
$OpDocDatasetTake
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset
Input dataset.
count
Number of elements to take.
name
Name for this dataset.
case class TensorDataset[T, O, D, S](data: T, name: String = "TensorDataset")(implicit evStructure: Aux[T, O, D, S], evData: Aux[T, O, D, S], evFunctionInput: ArgType[O]) extends Dataset[T, O, D, S] with Product with Serializable

Dataset with a single element.
Dataset with a single element.
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
data
Data representing the single element of this dataset.
name
Name for this dataset.
case class TensorSlicesDataset[T, O, D, S](data: T, name: String = "TensorSlicesDataset")(implicit evStructure: Aux[T, O, D, S], evData: Aux[T, O, D, S], evFunctionInput: ArgType[O]) extends Dataset[T, O, D, S] with Product with Serializable

Dataset with slices from the nested structure of Tensors (i.e., a Data-supported type).
Dataset with slices from the nested structure of Tensors (i.e., a Data-supported type). The slices are taken along the first axis of each Tensor in the nested structure.
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
data
Data representing the elements of this dataset.
name
Name for this dataset.
case class TextLinesDataset(filenames: tensors.Tensor[types.STRING], compressionType: CompressionType = NoCompression, bufferSize: Long = 256 * 1024, name: String = "TextLineDataset") extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape] with Product with Serializable

Dataset with elements read from text files (each line in each file corresponds to an element).
Dataset with elements read from text files (each line in each file corresponds to an element).
**Note:** New-line characters are stripped from the output.
filenames
Scalar or vector tensor containing the the name(s) of the file(s) to be read.
compressionType
Compression type for the file.
bufferSize
Number of bytes to buffer while reading from the file.
name
Name for this dataset.
case class Zip3Dataset[T1, O1, D1, S1, T2, O2, D2, S2, T3, O3, D3, S3](inputDataset1: Dataset[T1, O1, D1, S1], inputDataset2: Dataset[T2, O2, D2, S2], inputDataset3: Dataset[T3, O3, D3, S3], name: String = "Zip3Dataset")(implicit evStructure1: Aux[T1, O1, D1, S1] = inputDataset1.evStructure, evStructure2: Aux[T2, O2, D2, S2] = inputDataset2.evStructure, evStructure3: Aux[T3, O3, D3, S3] = inputDataset3.evStructure, evData1: Aux[T1, O1, D1, S1] = inputDataset1.evData, evData2: Aux[T2, O2, D2, S2] = inputDataset2.evData, evData3: Aux[T3, O3, D3, S3] = inputDataset3.evData, evFunctionInput1: ArgType[O1] = inputDataset1.evFunctionInput, evFunctionInput2: ArgType[O2] = inputDataset2.evFunctionInput, evFunctionInput3: ArgType[O3] = inputDataset3.evFunctionInput) extends Dataset[(T1, T2, T3), (O1, O2, O3), (D1, D2, D3), (S1, S2, S3)] with Product with Serializable

Dataset that wraps the application of the zip3 op.
Dataset that wraps the application of the zip3 op.
$OpDocDatasetZip
T1
First tensor type (i.e., nested structure of tensors).
O1
First output type (i.e., nested structure of symbolic tensors).
D1
First data type of the outputs (i.e., nested structure of TensorFlow data types).
S1
First shape type of the outputs (i.e., nested structure of TensorFlow shapes).
T2
Second tensor type (i.e., nested structure of tensors).
O2
Second output type (i.e., nested structure of symbolic tensors).
D2
Second data type of the outputs (i.e., nested structure of TensorFlow data types).
S2
Second shape type of the outputs (i.e., nested structure of TensorFlow shapes).
T3
Third tensor type (i.e., nested structure of tensors).
O3
Third output type (i.e., nested structure of symbolic tensors).
D3
Third data type of the outputs (i.e., nested structure of TensorFlow data types).
S3
Third shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset1
First input dataset.
inputDataset2
Second input dataset.
inputDataset3
Third input dataset.
name
Name for this dataset.
case class ZipDataset[T1, O1, D1, S1, T2, O2, D2, S2](inputDataset1: Dataset[T1, O1, D1, S1], inputDataset2: Dataset[T2, O2, D2, S2], name: String = "ZipDataset")(implicit evStructure1: Aux[T1, O1, D1, S1] = inputDataset1.evStructure, evStructure2: Aux[T2, O2, D2, S2] = inputDataset2.evStructure, evData1: Aux[T1, O1, D1, S1] = inputDataset1.evData, evData2: Aux[T2, O2, D2, S2] = inputDataset2.evData, evFunctionInput1: ArgType[O1] = inputDataset1.evFunctionInput, evFunctionInput2: ArgType[O2] = inputDataset2.evFunctionInput) extends Dataset[(T1, T2), (O1, O2), (D1, D2), (S1, S2)] with Product with Serializable

Dataset that wraps the application of the zip op.
Dataset that wraps the application of the zip op.
$OpDocDatasetZip
T1
First tensor type (i.e., nested structure of tensors).
O1
First output type (i.e., nested structure of symbolic tensors).
D1
First data type of the outputs (i.e., nested structure of TensorFlow data types).
S1
First shape type of the outputs (i.e., nested structure of TensorFlow shapes).
T2
Second tensor type (i.e., nested structure of tensors).
O2
Second output type (i.e., nested structure of symbolic tensors).
D2
Second data type of the outputs (i.e., nested structure of TensorFlow data types).
S2
Second shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDataset1
First input dataset.
inputDataset2
Second input dataset.
name
Name for this dataset.
case class ZipMultipleDataset[T, O, D, S](inputDatasets: Seq[Dataset[T, O, D, S]], name: String = "ZipMultipleDataset")(implicit evStructure: Aux[T, O, D, S] = inputDatasets.head.evStructure, evData: Aux[T, O, D, S] = inputDatasets.head.evData, evFunctionInput: ArgType[O] = inputDatasets.head.evFunctionInput, evFunctionInputSeq: ArgType[Seq[O]]) extends Dataset[Seq[T], Seq[O], Seq[D], Seq[S]] with Product with Serializable

Dataset that wraps the application of the zipMultiple op.
Dataset that wraps the application of the zipMultiple op.
$OpDocDatasetZip
T
Tensor type (i.e., nested structure of tensors).
O
Output type (i.e., nested structure of symbolic tensors).
D
Data type of the outputs (i.e., nested structure of TensorFlow data types).
S
Shape type of the outputs (i.e., nested structure of TensorFlow shapes).
inputDatasets
Input datasets.
name
Name for this dataset.

Value Members

object BatchDataset extends Serializable
object CacheDataset extends Serializable
object ConcatenatedDataset extends Serializable
object Data
object Dataset
object DropDataset extends Serializable
object DynamicTFRecordDataset
object FilterDataset extends Serializable
object FlatMapDataset extends Serializable
object GroupByWindowDataset extends Serializable
object IgnoreErrorsDataset extends Serializable
object InterleaveDataset extends Serializable
object Iterator

Contains helper functions for creating iterator-related ops, as well as the iterator API trait.
object MapDataset extends Serializable
object PaddedBatchDataset extends Serializable
object PrefetchDataset extends Serializable
object RangeDataset extends Serializable
object RepeatDataset extends Serializable
object ShuffleDataset extends Serializable
object TFRecordDataset
object TakeDataset extends Serializable
object ZipDataset extends Serializable

package data

Type Members

case class BatchDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], batchSize: Long, name: String = "BatchDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class CacheDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], directory: String, name: String = "CacheDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class ConcatenatedDataset[T, O, D, S](inputDataset1: Dataset[T, O, D, S], inputDataset2: Dataset[T, O, D, S], name: String = "ConcatenatedDataset") extends Dataset[T, O, D, S] with Product with Serializable

trait Data[T] extends AnyRef

abstract class Dataset[T, O, D, S] extends AnyRef

case class DropDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Long, name: String = "DropDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class DynamicDropDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Output, name: String = "DropDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class DynamicPaddedBatchDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], batchSize: Long, paddedShapes: S, paddingValues: O = null.asInstanceOf[O], name: String = "PaddedBatchDataset") extends Dataset[T, O, D, S] with Product with Serializable

class DynamicTFRecordDataset extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape]

case class DynamicTakeDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Output, name: String = "TakeDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class DynamicTextLinesDataset(filenames: Output, compressionType: CompressionType = NoCompression, bufferSize: Long = 256 * 1024, name: String = "TextLineDataset") extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape] with Product with Serializable

case class FilterDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], predicateFn: (O) ⇒ Output, name: String = "FilterDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class IgnoreErrorsDataset[T, O, D, S] extends Dataset[T, O, D, S] with Product with Serializable

class InitializableIterator[T, O, D, S] extends Iterator[T, O, D, S]

class Iterator[T, O, D, S] extends AnyRef

case class OutputDataset[T, O, D, S](data: O, name: String = "OutputDataset")(implicit evStructure: Aux[T, O, D, S], evData: Aux[T, O, D, S], evFunctionInput: ArgType[O]) extends Dataset[T, O, D, S] with Product with Serializable

case class OutputSlicesDataset[T, O, D, S](data: O, name: String = "OutputSlicesDataset")(implicit evStructure: Aux[T, O, D, S], evData: Aux[T, O, D, S], evFunctionInput: ArgType[O]) extends Dataset[T, O, D, S] with Product with Serializable

case class PaddedBatchDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], batchSize: Long, paddedShapes: S, paddingValues: T = null.asInstanceOf[T], name: String = "PaddedBatchDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class PrefetchDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], bufferSize: Long, name: String = "PrefetchDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class RangeDataset(start: Long, limit: Long, delta: Long = 1L, name: String = "RangeDataset") extends Dataset[tensors.Tensor[types.INT64], Output, types.INT64, core.Shape] with Product with Serializable

case class RepeatDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Long, name: String = "RepeatDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class ShuffleDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], bufferSize: Long, seed: Option[Int], name: String = "ShuffleDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class SparseOutputSlicesDataset(tensor: SparseOutput, name: String = "SparseOutputSliceDataset") extends Dataset[tensors.SparseTensor[types.DataType], SparseOutput, (types.INT64, types.DataType, types.INT64), (core.Shape, core.Shape, core.Shape)] with Product with Serializable

case class SparseTensorSlicesDataset[D <: types.DataType](tensor: tensors.SparseTensor[D], name: String = "SparseTensorSliceDataset") extends Dataset[tensors.SparseTensor[D], SparseOutput, (types.INT64, D, types.INT64), (core.Shape, core.Shape, core.Shape)] with Product with Serializable

class TFRecordDataset extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape]

case class TakeDataset[T, O, D, S](inputDataset: Dataset[T, O, D, S], count: Long, name: String = "TakeDataset") extends Dataset[T, O, D, S] with Product with Serializable

case class TensorDataset[T, O, D, S](data: T, name: String = "TensorDataset")(implicit evStructure: Aux[T, O, D, S], evData: Aux[T, O, D, S], evFunctionInput: ArgType[O]) extends Dataset[T, O, D, S] with Product with Serializable

case class TensorSlicesDataset[T, O, D, S](data: T, name: String = "TensorSlicesDataset")(implicit evStructure: Aux[T, O, D, S], evData: Aux[T, O, D, S], evFunctionInput: ArgType[O]) extends Dataset[T, O, D, S] with Product with Serializable

case class TextLinesDataset(filenames: tensors.Tensor[types.STRING], compressionType: CompressionType = NoCompression, bufferSize: Long = 256 * 1024, name: String = "TextLineDataset") extends Dataset[tensors.Tensor[types.STRING], Output, types.STRING, core.Shape] with Product with Serializable

Value Members

object BatchDataset extends Serializable

object CacheDataset extends Serializable

object ConcatenatedDataset extends Serializable

object Data

object Dataset

object DropDataset extends Serializable

object DynamicTFRecordDataset

object FilterDataset extends Serializable

object FlatMapDataset extends Serializable

object GroupByWindowDataset extends Serializable

object IgnoreErrorsDataset extends Serializable

object InterleaveDataset extends Serializable

object Iterator

object MapDataset extends Serializable

object PaddedBatchDataset extends Serializable

object PrefetchDataset extends Serializable

object RangeDataset extends Serializable

object RepeatDataset extends Serializable

object ShuffleDataset extends Serializable

object TFRecordDataset

object TakeDataset extends Serializable

object ZipDataset extends Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped