org.bytedeco.javacpp

Class tensorflow.Dequantize

java.lang.Object

org.bytedeco.javacpp.Pointer

org.bytedeco.javacpp.tensorflow.Dequantize

All Implemented Interfaces:

AutoCloseable

Enclosing class:

tensorflow

@Namespace(value="tensorflow::ops")
 @NoOffset
public static class tensorflow.Dequantize
extends Pointer

Dequantize the 'input' tensor into a float Tensor. [min_range, max_range] are scalar floats that specify the range for the 'input' data. The 'mode' attribute controls exactly which calculations are used to convert the float values to their quantized equivalents. In 'MIN_COMBINED' mode, each value of the tensor will undergo the following:

  if T == qint8, in[i] += (range(T) + 1)/ 2.0
  out[i] = min_range + (in[i]* (max_range - min_range) / range(T))
  

here range(T) = numeric_limits<T>::max() - numeric_limits<T>::min() *MIN_COMBINED Mode Example* If the input comes from a QuantizedRelu6, the output type is quint8 (range of 0-255) but the possible range of QuantizedRelu6 is 0-6. The min_range and max_range values are therefore 0.0 and 6.0. Dequantize on quint8 will take each value, cast to float, and multiply by 6 / 255. Note that if quantizedtype is qint8, the operation will additionally add each value by 128 prior to casting. If the mode is 'MIN_FIRST', then this approach is used:

 c++
  num_discrete_values = 1 << (# of bits in T)
  range_adjust = num_discrete_values / (num_discrete_values - 1)
  range = (range_max - range_min) * range_adjust
  range_scale = range / num_discrete_values
  const double offset_input = static_cast<double>(input) - lowest_quantized;
  result = range_min + ((input - numeric_limits<T>::min()) * range_scale)
  

*SCALED mode Example* SCALED mode matches the quantization approach used in QuantizeAndDequantize{V2|V3}. If the mode is SCALED, we do not use the full range of the output type, choosing to elide the lowest possible value for symmetry (e.g., output range is -127 to 127, not -128 to 127 for signed 8 bit quantization), so that 0.0 maps to 0. We first find the range of values in our tensor. The range we use is always centered on 0, so we find m such that

 c++
    m = max(abs(input_min), abs(input_max))
  

Our input tensor range is then [-m, m]. Next, we choose our fixed-point quantization buckets, [min_fixed, max_fixed]. If T is signed, this is

    num_bits = sizeof(T) * 8
    [min_fixed, max_fixed] =
        [-(1 << (num_bits - 1) - 1), (1 << (num_bits - 1)) - 1]
  

Otherwise, if T is unsigned, the fixed-point range is

    [min_fixed, max_fixed] = [0, (1 << num_bits) - 1]
  

From this we compute our scaling factor, s:

 c++
    s = (2 * m) / (max_fixed - min_fixed)
  

Now we can dequantize the elements of our tensor:

 c++
  result = input * s
  

Arguments: * scope: A Scope object * min_range: The minimum scalar value possibly produced for the input. * max_range: The maximum scalar value possibly produced for the input. Returns: * Output: The output tensor.

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	tensorflow.Dequantize.Attrs Optional attribute setters for Dequantize

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

Pointer.CustomDeallocator, Pointer.Deallocator, Pointer.NativeDeallocator

Field Summary

Fields inherited from class org.bytedeco.javacpp.Pointer

address, capacity, limit, position

Constructor Summary

Constructors

Constructor and Description
Dequantize(Pointer p) Pointer cast constructor.
Dequantize(tensorflow.Scope scope, tensorflow.Input input, tensorflow.Input min_range, tensorflow.Input max_range)
Dequantize(tensorflow.Scope scope, tensorflow.Input input, tensorflow.Input min_range, tensorflow.Input max_range, tensorflow.Dequantize.Attrs attrs)

Method Summary

Modifier and Type	Method and Description
tensorflow.Input	asInput()
tensorflow.Output	asOutput()
static tensorflow.Dequantize.Attrs	Mode(BytePointer x)
static tensorflow.Dequantize.Attrs	Mode(String x)
tensorflow.Node	node()
tensorflow.Operation	operation()
tensorflow.Dequantize	operation(tensorflow.Operation operation)
tensorflow.Output	output()
tensorflow.Dequantize	output(tensorflow.Output output)

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, hashCode, isNull, limit, limit, malloc, maxBytes, maxPhysicalBytes, memchr, memcmp, memcpy, memmove, memset, offsetof, parseBytes, physicalBytes, position, position, put, realloc, setNull, sizeof, toString, totalBytes, totalPhysicalBytes, withDeallocator, zero

Methods inherited from class java.lang.Object

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

Constructor Detail

Dequantize

public Dequantize(Pointer p)

Pointer cast constructor. Invokes Pointer.Pointer(Pointer).

Dequantize

public Dequantize(@Const @ByRef
                  tensorflow.Scope scope,
                  @ByVal
                  tensorflow.Input input,
                  @ByVal
                  tensorflow.Input min_range,
                  @ByVal
                  tensorflow.Input max_range)

Dequantize

public Dequantize(@Const @ByRef
                  tensorflow.Scope scope,
                  @ByVal
                  tensorflow.Input input,
                  @ByVal
                  tensorflow.Input min_range,
                  @ByVal
                  tensorflow.Input max_range,
                  @Const @ByRef
                  tensorflow.Dequantize.Attrs attrs)

Method Detail

asOutput

@ByVal
 @Name(value="operator tensorflow::Output")
public tensorflow.Output asOutput()

asInput

@ByVal
 @Name(value="operator tensorflow::Input")
public tensorflow.Input asInput()

node

public tensorflow.Node node()

Mode

@ByVal
public static tensorflow.Dequantize.Attrs Mode(@tensorflow.StringPiece
                                                       BytePointer x)

Mode

@ByVal
public static tensorflow.Dequantize.Attrs Mode(@tensorflow.StringPiece
                                                       String x)

operation

@ByRef
public tensorflow.Operation operation()

operation

public tensorflow.Dequantize operation(tensorflow.Operation operation)

output

@ByRef
public tensorflow.Output output()

output

public tensorflow.Dequantize output(tensorflow.Output output)