Creates a dataset that emits the key-value pairs in one or more LMDB files.
The Lightning Memory-Mapped Database Manager, or LMDB, is an embedded binary key-value database. This dataset can read the contents of LMDB database files, the names of which generally have the `.mdb` suffix.
Each output element consists of a key-value pair represented as a pair of scalar string `Tensor`s, where the first `Tensor` contains the key and the second `Tensor` contains the value.
LMDB uses different file formats on big- and little-endian machines. `data.LMDBDataset` can only read files in the format of the host machine.
Constants
| String | OP_NAME | The name of this op, as known by TensorFlow core engine |
Public Methods
| Output<TType> |
asOutput()
Returns the symbolic handle of the tensor.
|
| static LMDBDataset | |
| Output<?> |
handle()
|
Inherited Methods
Constants
public static final String OP_NAME
The name of this op, as known by TensorFlow core engine
Public Methods
public Output<TType> asOutput ()
Returns the symbolic handle of the 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 LMDBDataset create (Scope scope, Operand<TString> filenames, List<Class<? extends TType>> outputTypes, List<Shape> outputShapes)
Factory method to create a class wrapping a new LMDBDataset operation.
Parameters
| scope | current scope |
|---|---|
| filenames | A scalar or a vector containing the name(s) of the binary file(s) to be read. |
Returns
- a new instance of LMDBDataset