org.bytedeco.tensorflowlite

Class SignatureRunner

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.tensorflowlite.SignatureRunner

All Implemented Interfaces:

AutoCloseable

@Namespace(value="tflite::impl")
 @NoOffset
 @Properties(inherit=tensorflowlite.class)
public class SignatureRunner
extends Pointer

runner->Invoke(); /** /** /** WARNING: This class is *not* thread-safe. The client is responsible for /** ensuring serialized interaction to avoid data races and undefined behavior. /** /** SignatureRunner and Interpreter share the same underlying data. Calling /** methods on an Interpreter object will affect the state in corresponding /** SignatureRunner objects. Therefore, it is recommended not to call other /** Interpreter methods after calling GetSignatureRunner to create /** SignatureRunner instances.

Nested Class Summary

Nested classes/interfaces inherited from class org.bytedeco.javacpp.Pointer

Pointer.CustomDeallocator, Pointer.Deallocator, Pointer.NativeDeallocator, Pointer.ReferenceCounter

Field Summary

Fields inherited from class org.bytedeco.javacpp.Pointer

address, capacity, limit, position

Constructor Summary

Constructors

Constructor and Description
SignatureRunner(Pointer p) Pointer cast constructor.

Method Summary

Modifier and Type	Method and Description
int	AllocateTensors() Updates allocations for all tensors, related to the given signature.
int	Cancel() Attempts to cancel in flight invocation if any.
PointerPointer	input_names() Read-only access to list of signature input names.
long	input_size() Returns the number of inputs.
TfLiteTensor	input_tensor(BytePointer input_name) Returns the input tensor identified by 'input_name' in the given signature.
TfLiteTensor	input_tensor(String input_name)
int	Invoke() Invokes the signature runner (run the graph identified by the given signature in dependency order).
PointerPointer	output_names() Read-only access to list of signature output names.
long	output_size() Returns the number of outputs.
TfLiteTensor	output_tensor(BytePointer output_name) Returns the output tensor identified by 'output_name' in the given signature.
TfLiteTensor	output_tensor(String output_name)
int	ResizeInputTensor(BytePointer input_name, int[] new_size)
int	ResizeInputTensor(BytePointer input_name, IntBuffer new_size)
int	ResizeInputTensor(BytePointer input_name, IntPointer new_size) Change a dimensionality of a given tensor.
int	ResizeInputTensor(String input_name, int[] new_size)
int	ResizeInputTensor(String input_name, IntBuffer new_size)
int	ResizeInputTensor(String input_name, IntPointer new_size)
int	ResizeInputTensorStrict(BytePointer input_name, int[] new_size)
int	ResizeInputTensorStrict(BytePointer input_name, IntBuffer new_size)
int	ResizeInputTensorStrict(BytePointer input_name, IntPointer new_size) Change the dimensionality of a given tensor.
int	ResizeInputTensorStrict(String input_name, int[] new_size)
int	ResizeInputTensorStrict(String input_name, IntBuffer new_size)
int	ResizeInputTensorStrict(String input_name, IntPointer new_size)
void	SetAllowBufferHandleOutput(boolean allow_buffer_handle_output) \brief Set if buffer handle output is allowed.
int	SetCustomAllocationForInputTensor(BytePointer input_name, TfLiteCustomAllocation allocation)
int	SetCustomAllocationForInputTensor(BytePointer input_name, TfLiteCustomAllocation allocation, long flags) \brief Assigns (or reassigns) a custom memory allocation for the given tensor name.
int	SetCustomAllocationForInputTensor(String input_name, TfLiteCustomAllocation allocation)
int	SetCustomAllocationForInputTensor(String input_name, TfLiteCustomAllocation allocation, long flags)
int	SetCustomAllocationForOutputTensor(BytePointer output_name, TfLiteCustomAllocation allocation)
int	SetCustomAllocationForOutputTensor(BytePointer output_name, TfLiteCustomAllocation allocation, long flags) \brief Assigns (or reassigns) a custom memory allocation for the given tensor name.
int	SetCustomAllocationForOutputTensor(String output_name, TfLiteCustomAllocation allocation)
int	SetCustomAllocationForOutputTensor(String output_name, TfLiteCustomAllocation allocation, long flags)
int	SetInputBufferHandle(BytePointer input_name, int buffer_handle, TfLiteDelegate delegate)
int	SetInputBufferHandle(BytePointer input_name, int buffer_handle, TfLiteDelegate delegate, boolean release_existing_buffer_handle) \warning This is an experimental API and subject to change.
int	SetInputBufferHandle(String input_name, int buffer_handle, TfLiteDelegate delegate)
int	SetInputBufferHandle(String input_name, int buffer_handle, TfLiteDelegate delegate, boolean release_existing_buffer_handle)
int	SetOutputBufferHandle(BytePointer output_name, int buffer_handle, TfLiteDelegate delegate)
int	SetOutputBufferHandle(BytePointer output_name, int buffer_handle, TfLiteDelegate delegate, boolean release_existing_buffer_handle) \warning This is an experimental API and subject to change.
int	SetOutputBufferHandle(String output_name, int buffer_handle, TfLiteDelegate delegate)
int	SetOutputBufferHandle(String output_name, int buffer_handle, TfLiteDelegate delegate, boolean release_existing_buffer_handle)
String	signature_key() Returns the key for the corresponding signature.

Methods inherited from class org.bytedeco.javacpp.Pointer

address, asBuffer, asByteBuffer, availablePhysicalBytes, calloc, capacity, capacity, close, deallocate, deallocate, deallocateReferences, deallocator, deallocator, equals, fill, formatBytes, free, getDirectBufferAddress, getPointer, getPointer, getPointer, getPointer, hashCode, interruptDeallocatorThread, isNull, isNull, limit, limit, malloc, maxBytes, maxPhysicalBytes, memchr, memcmp, memcpy, memmove, memset, offsetAddress, offsetof, offsetof, parseBytes, physicalBytes, physicalBytesInaccurate, position, position, put, realloc, referenceCount, releaseReference, retainReference, setNull, sizeof, sizeof, toString, totalBytes, totalCount, totalPhysicalBytes, withDeallocator, zero

Methods inherited from class java.lang.Object

clone, finalize, getClass, notify, notifyAll, wait, wait, wait

Constructor Detail

SignatureRunner

public SignatureRunner(Pointer p)

Pointer cast constructor. Invokes Pointer(Pointer).

Method Detail

signature_key

@StdString
public String signature_key()

Returns the key for the corresponding signature.

input_size

@Cast(value="size_t")
public long input_size()

Returns the number of inputs.

output_size

@Cast(value="size_t")
public long output_size()

Returns the number of outputs.

input_names

@Cast(value="const char**")
 @StdVector
public PointerPointer input_names()

Read-only access to list of signature input names.

output_names

@Cast(value="const char**")
 @StdVector
public PointerPointer output_names()

Read-only access to list of signature output names.

input_tensor

public TfLiteTensor input_tensor(@Cast(value="const char*")
                                 BytePointer input_name)

Returns the input tensor identified by 'input_name' in the given signature. Returns nullptr if the given name is not valid.

input_tensor

public TfLiteTensor input_tensor(String input_name)

output_tensor

@Const
public TfLiteTensor output_tensor(@Cast(value="const char*")
                                         BytePointer output_name)

Returns the output tensor identified by 'output_name' in the given signature. Returns nullptr if the given name is not valid.

output_tensor

@Const
public TfLiteTensor output_tensor(String output_name)

ResizeInputTensor

@Cast(value="TfLiteStatus")
public int ResizeInputTensor(@Cast(value="const char*")
                                                         BytePointer input_name,
                                                         @StdVector
                                                         IntPointer new_size)

Change a dimensionality of a given tensor. Note, this is only acceptable for tensors that are inputs. Returns status of failure or success. Note that this doesn't actually resize any existing buffers. A call to AllocateTensors() is required to change the tensor input buffer.

ResizeInputTensor

@Cast(value="TfLiteStatus")
public int ResizeInputTensor(String input_name,
                                                         @StdVector
                                                         IntBuffer new_size)

ResizeInputTensor

@Cast(value="TfLiteStatus")
public int ResizeInputTensor(@Cast(value="const char*")
                                                         BytePointer input_name,
                                                         @StdVector
                                                         int[] new_size)

ResizeInputTensor

@Cast(value="TfLiteStatus")
public int ResizeInputTensor(String input_name,
                                                         @StdVector
                                                         IntPointer new_size)

ResizeInputTensor

@Cast(value="TfLiteStatus")
public int ResizeInputTensor(@Cast(value="const char*")
                                                         BytePointer input_name,
                                                         @StdVector
                                                         IntBuffer new_size)

ResizeInputTensor

@Cast(value="TfLiteStatus")
public int ResizeInputTensor(String input_name,
                                                         @StdVector
                                                         int[] new_size)

ResizeInputTensorStrict

@Cast(value="TfLiteStatus")
public int ResizeInputTensorStrict(@Cast(value="const char*")
                                                               BytePointer input_name,
                                                               @StdVector
                                                               IntPointer new_size)

Change the dimensionality of a given tensor. This is only acceptable for tensor indices that are inputs or variables. Difference from ResizeInputTensor: Only unknown dimensions can be resized with this function. Unknown dimensions are indicated as -1 in the dims_signature attribute of a TfLiteTensor. Returns status of failure or success. Note that this doesn't actually resize any existing buffers. A call to AllocateTensors() is required to change the tensor input buffer.

ResizeInputTensorStrict

@Cast(value="TfLiteStatus")
public int ResizeInputTensorStrict(String input_name,
                                                               @StdVector
                                                               IntBuffer new_size)

ResizeInputTensorStrict

@Cast(value="TfLiteStatus")
public int ResizeInputTensorStrict(@Cast(value="const char*")
                                                               BytePointer input_name,
                                                               @StdVector
                                                               int[] new_size)

ResizeInputTensorStrict

@Cast(value="TfLiteStatus")
public int ResizeInputTensorStrict(String input_name,
                                                               @StdVector
                                                               IntPointer new_size)

ResizeInputTensorStrict

@Cast(value="TfLiteStatus")
public int ResizeInputTensorStrict(@Cast(value="const char*")
                                                               BytePointer input_name,
                                                               @StdVector
                                                               IntBuffer new_size)

ResizeInputTensorStrict

@Cast(value="TfLiteStatus")
public int ResizeInputTensorStrict(String input_name,
                                                               @StdVector
                                                               int[] new_size)

AllocateTensors

@Cast(value="TfLiteStatus")
public int AllocateTensors()

Updates allocations for all tensors, related to the given signature.

Invoke

@Cast(value="TfLiteStatus")
public int Invoke()

Invokes the signature runner (run the graph identified by the given signature in dependency order).

Cancel

@Cast(value="TfLiteStatus")
public int Cancel()

Attempts to cancel in flight invocation if any. This will not affect calls to Invoke that happened after this. Non blocking and thread safe. Returns kTfLiteError if cancellation is not enabled, otherwise returns kTfLiteOk. WARNING: This is an experimental API and subject to change.

SetCustomAllocationForInputTensor

@Cast(value="TfLiteStatus")
public int SetCustomAllocationForInputTensor(@Cast(value="const char*")
                                                                         BytePointer input_name,
                                                                         @Const @ByRef
                                                                         TfLiteCustomAllocation allocation,
                                                                         @Cast(value="int64_t")
                                                                         long flags)

\brief Assigns (or reassigns) a custom memory allocation for the given tensor name. flags is a bitmask, see TfLiteCustomAllocationFlags. The runtime does NOT take ownership of the underlying memory. NOTE: User needs to call AllocateTensors() after this. Invalid/insufficient buffers will cause an error during AllocateTensors or Invoke (in case of dynamic shapes in the graph). Parameters should satisfy the following conditions: 1. tensor->allocation_type == kTfLiteArenaRw or kTfLiteArenaRwPersistent In general, this is true for I/O tensors & variable tensors. 2. allocation->data has the appropriate permissions for runtime access (Read-only for inputs, Read-Write for others), and outlives Interpreter. 3. allocation->bytes >= tensor->bytes. This condition is checked again if any tensors are resized. 4. allocation->data should be aligned to kDefaultTensorAlignment defined in lite/util.h. (Currently 64 bytes) This check is skipped if kTfLiteCustomAllocationFlagsSkipAlignCheck is set through flags. \warning This is an experimental API and subject to change. \n

SetCustomAllocationForInputTensor

@Cast(value="TfLiteStatus")
public int SetCustomAllocationForInputTensor(@Cast(value="const char*")
                                                                         BytePointer input_name,
                                                                         @Const @ByRef
                                                                         TfLiteCustomAllocation allocation)

SetCustomAllocationForInputTensor

@Cast(value="TfLiteStatus")
public int SetCustomAllocationForInputTensor(String input_name,
                                                                         @Const @ByRef
                                                                         TfLiteCustomAllocation allocation,
                                                                         @Cast(value="int64_t")
                                                                         long flags)

SetCustomAllocationForInputTensor

@Cast(value="TfLiteStatus")
public int SetCustomAllocationForInputTensor(String input_name,
                                                                         @Const @ByRef
                                                                         TfLiteCustomAllocation allocation)

SetCustomAllocationForOutputTensor

@Cast(value="TfLiteStatus")
public int SetCustomAllocationForOutputTensor(@Cast(value="const char*")
                                                                          BytePointer output_name,
                                                                          @Const @ByRef
                                                                          TfLiteCustomAllocation allocation,
                                                                          @Cast(value="int64_t")
                                                                          long flags)

\brief Assigns (or reassigns) a custom memory allocation for the given tensor name. flags is a bitmask, see TfLiteCustomAllocationFlags. The runtime does NOT take ownership of the underlying memory. NOTE: User needs to call AllocateTensors() after this. Invalid/insufficient buffers will cause an error during AllocateTensors or Invoke (in case of dynamic shapes in the graph). Parameters should satisfy the following conditions: 1. tensor->allocation_type == kTfLiteArenaRw or kTfLiteArenaRwPersistent In general, this is true for I/O tensors & variable tensors. 2. allocation->data has the appropriate permissions for runtime access (Read-only for inputs, Read-Write for others), and outlives Interpreter. 3. allocation->bytes >= tensor->bytes. This condition is checked again if any tensors are resized. 4. allocation->data should be aligned to kDefaultTensorAlignment defined in lite/util.h. (Currently 64 bytes) This check is skipped if kTfLiteCustomAllocationFlagsSkipAlignCheck is set through flags. \warning This is an experimental API and subject to change. \n

SetCustomAllocationForOutputTensor

@Cast(value="TfLiteStatus")
public int SetCustomAllocationForOutputTensor(@Cast(value="const char*")
                                                                          BytePointer output_name,
                                                                          @Const @ByRef
                                                                          TfLiteCustomAllocation allocation)

SetCustomAllocationForOutputTensor

@Cast(value="TfLiteStatus")
public int SetCustomAllocationForOutputTensor(String output_name,
                                                                          @Const @ByRef
                                                                          TfLiteCustomAllocation allocation,
                                                                          @Cast(value="int64_t")
                                                                          long flags)

SetCustomAllocationForOutputTensor

@Cast(value="TfLiteStatus")
public int SetCustomAllocationForOutputTensor(String output_name,
                                                                          @Const @ByRef
                                                                          TfLiteCustomAllocation allocation)

SetAllowBufferHandleOutput

public void SetAllowBufferHandleOutput(@Cast(value="bool")
                                       boolean allow_buffer_handle_output)

\brief Set if buffer handle output is allowed. When using hardware delegation, Interpreter will make the data of output tensors available in tensor->data by default. If the application can consume the buffer handle directly (e.g. reading output from OpenGL texture), it can set this flag to true, so Interpreter won't copy the data from buffer handle to CPU memory. \warning This is an experimental API and subject to change. \n

SetInputBufferHandle

@Cast(value="TfLiteStatus")
public int SetInputBufferHandle(@Cast(value="const char*")
                                                            BytePointer input_name,
                                                            @Cast(value="TfLiteBufferHandle")
                                                            int buffer_handle,
                                                            TfLiteDelegate delegate,
                                                            @Cast(value="bool")
                                                            boolean release_existing_buffer_handle)

\warning This is an experimental API and subject to change. \n \brief Set the delegate buffer handle to a input tensor. TfLiteDelegate should be aware of how to handle the buffer handle. release_existing_buffer_handle: If true, the existing buffer handle

SetInputBufferHandle

@Cast(value="TfLiteStatus")
public int SetInputBufferHandle(@Cast(value="const char*")
                                                            BytePointer input_name,
                                                            @Cast(value="TfLiteBufferHandle")
                                                            int buffer_handle,
                                                            TfLiteDelegate delegate)

SetInputBufferHandle

@Cast(value="TfLiteStatus")
public int SetInputBufferHandle(String input_name,
                                                            @Cast(value="TfLiteBufferHandle")
                                                            int buffer_handle,
                                                            TfLiteDelegate delegate,
                                                            @Cast(value="bool")
                                                            boolean release_existing_buffer_handle)

SetInputBufferHandle

@Cast(value="TfLiteStatus")
public int SetInputBufferHandle(String input_name,
                                                            @Cast(value="TfLiteBufferHandle")
                                                            int buffer_handle,
                                                            TfLiteDelegate delegate)

SetOutputBufferHandle

@Cast(value="TfLiteStatus")
public int SetOutputBufferHandle(@Cast(value="const char*")
                                                             BytePointer output_name,
                                                             @Cast(value="TfLiteBufferHandle")
                                                             int buffer_handle,
                                                             TfLiteDelegate delegate,
                                                             @Cast(value="bool")
                                                             boolean release_existing_buffer_handle)

\warning This is an experimental API and subject to change. \n \brief Set the delegate buffer handle to a output tensor. TfLiteDelegate should be aware of how to handle the buffer handle. release_existing_buffer_handle: If true, the existing buffer handle will be released by TfLiteDelegate::FreeBufferHandle.

SetOutputBufferHandle

@Cast(value="TfLiteStatus")
public int SetOutputBufferHandle(@Cast(value="const char*")
                                                             BytePointer output_name,
                                                             @Cast(value="TfLiteBufferHandle")
                                                             int buffer_handle,
                                                             TfLiteDelegate delegate)

SetOutputBufferHandle

@Cast(value="TfLiteStatus")
public int SetOutputBufferHandle(String output_name,
                                                             @Cast(value="TfLiteBufferHandle")
                                                             int buffer_handle,
                                                             TfLiteDelegate delegate,
                                                             @Cast(value="bool")
                                                             boolean release_existing_buffer_handle)

SetOutputBufferHandle

@Cast(value="TfLiteStatus")
public int SetOutputBufferHandle(String output_name,
                                                             @Cast(value="TfLiteBufferHandle")
                                                             int buffer_handle,
                                                             TfLiteDelegate delegate)