ai.djl.pytorch.engine

Class PtSymbolBlock

java.lang.Object

ai.djl.nn.AbstractBlock

ai.djl.nn.AbstractSymbolBlock

ai.djl.pytorch.engine.PtSymbolBlock

All Implemented Interfaces:

ai.djl.nn.Block, ai.djl.nn.SymbolBlock, java.lang.AutoCloseable

public class PtSymbolBlock
extends ai.djl.nn.AbstractSymbolBlock
implements java.lang.AutoCloseable

PtSymbolBlock is the PyTorch implementation of SymbolBlock.

You can create a PtSymbolBlock using Model.load(java.nio.file.Path, String).

Field Summary

Fields inherited from class ai.djl.nn.AbstractBlock

children, inputNames, inputShapes, parameters, version

Constructor Summary

Constructors

Constructor and Description
PtSymbolBlock(PtNDManager manager) Constructs an Empty PtSymbolBlock.
PtSymbolBlock(PtNDManager manager, long handle) Constructs a PtSymbolBlock.

Method Summary

Modifier and Type	Method and Description
void	close()
ai.djl.util.PairList<java.lang.String,ai.djl.ndarray.types.Shape>	describeInput()
ai.djl.util.PairList<java.lang.String,ai.djl.ndarray.types.Shape>	describeOutput()
IValue	forward(IValue... inputs) Runs the forward of this PyTorch module.
protected ai.djl.ndarray.NDList	forwardInternal(ai.djl.training.ParameterStore parameterStore, ai.djl.ndarray.NDList inputs, boolean training, ai.djl.util.PairList<java.lang.String,java.lang.Object> params)
java.lang.Long	getHandle() Get the native PyTorch model pointer.
ai.djl.ndarray.types.Shape[]	getOutputShapes(ai.djl.ndarray.types.Shape[] inputShapes)
void	loadParameters(ai.djl.ndarray.NDManager manager, java.io.DataInputStream is)
void	saveParameters(java.io.DataOutputStream os)

Methods inherited from class ai.djl.nn.AbstractBlock

addChildBlock, addParameter, beforeInitialize, cast, clear, forward, forward, forwardInternal, getChildren, getDirectParameters, getParameters, initialize, initializeChildBlocks, isInitialized, loadMetadata, prepare, readInputShapes, saveInputShapes, saveMetadata, setInitializer, setInitializer, setInitializer, toString

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface ai.djl.nn.SymbolBlock

newInstance, removeLastBlock

Methods inherited from interface ai.djl.nn.Block

cast, clear, forward, forward, forward, getChildren, getDirectParameters, getParameters, initialize, isInitialized, setInitializer, setInitializer, setInitializer, validateLayout

Constructor Detail

PtSymbolBlock

public PtSymbolBlock(PtNDManager manager,
                     long handle)

Constructs a PtSymbolBlock.

You can create a PtSymbolBlock using Model.load(java.nio.file.Path, String).

Parameters:

manager - the manager to use for the block

handle - the module handle

PtSymbolBlock

public PtSymbolBlock(PtNDManager manager)

Constructs an Empty PtSymbolBlock.

Parameters:

manager - the manager to use for the block

Method Detail

close

public void close()

Specified by:

close in interface java.lang.AutoCloseable

forward

public IValue forward(IValue... inputs)

Runs the forward of this PyTorch module.

Parameters:

inputs - the input IValue

Returns:

the result IValue

forwardInternal

protected ai.djl.ndarray.NDList forwardInternal(ai.djl.training.ParameterStore parameterStore,
                                                ai.djl.ndarray.NDList inputs,
                                                boolean training,
                                                ai.djl.util.PairList<java.lang.String,java.lang.Object> params)

Specified by:

forwardInternal in class ai.djl.nn.AbstractBlock

describeInput

public ai.djl.util.PairList<java.lang.String,ai.djl.ndarray.types.Shape> describeInput()

Specified by:

describeInput in interface ai.djl.nn.Block

Overrides:

describeInput in class ai.djl.nn.AbstractBlock

describeOutput

public ai.djl.util.PairList<java.lang.String,ai.djl.ndarray.types.Shape> describeOutput()

Specified by:

describeOutput in interface ai.djl.nn.SymbolBlock

getOutputShapes

public ai.djl.ndarray.types.Shape[] getOutputShapes(ai.djl.ndarray.types.Shape[] inputShapes)

Specified by:

getOutputShapes in interface ai.djl.nn.Block

Overrides:

getOutputShapes in class ai.djl.nn.AbstractSymbolBlock

saveParameters

public void saveParameters(java.io.DataOutputStream os)
                    throws java.io.IOException

Specified by:

saveParameters in interface ai.djl.nn.Block

Overrides:

saveParameters in class ai.djl.nn.AbstractBlock

Throws:

java.io.IOException

loadParameters

public void loadParameters(ai.djl.ndarray.NDManager manager,
                           java.io.DataInputStream is)
                    throws java.io.IOException,
                           ai.djl.MalformedModelException

Specified by:

loadParameters in interface ai.djl.nn.Block

Overrides:

loadParameters in class ai.djl.nn.AbstractBlock

Throws:

java.io.IOException

ai.djl.MalformedModelException

getHandle

public java.lang.Long getHandle()

Get the native PyTorch model pointer.

Returns:

the pointer