org.tensorflow.op.quantization

Classes

Dequantize <U extends TNumber > Dequantize the 'input' tensor into a float or bfloat16 Tensor.
Dequantize.Options Optional attributes for Dequantize
FakeQuantWithMinMaxArgs Fake-quantize the 'inputs' tensor, type float to 'outputs' tensor of same type.
FakeQuantWithMinMaxArgs.Options Optional attributes for FakeQuantWithMinMaxArgs
FakeQuantWithMinMaxArgsGradient Compute gradients for a FakeQuantWithMinMaxArgs operation.
FakeQuantWithMinMaxArgsGradient.Options Optional attributes for FakeQuantWithMinMaxArgsGradient
FakeQuantWithMinMaxVars Fake-quantize the 'inputs' tensor of type float via global float scalars

Fake-quantize the `inputs` tensor of type float via global float scalars `min` and `max` to `outputs` tensor of same shape as `inputs`.

FakeQuantWithMinMaxVars.Options Optional attributes for FakeQuantWithMinMaxVars
FakeQuantWithMinMaxVarsGradient Compute gradients for a FakeQuantWithMinMaxVars operation.
FakeQuantWithMinMaxVarsGradient.Options Optional attributes for FakeQuantWithMinMaxVarsGradient
FakeQuantWithMinMaxVarsPerChannel Fake-quantize the 'inputs' tensor of type float via per-channel floats

Fake-quantize the `inputs` tensor of type float per-channel and one of the shapes: `[d]`, `[b, d]` `[b, h, w, d]` via per-channel floats `min` and `max` of shape `[d]` to `outputs` tensor of same shape as `inputs`.

FakeQuantWithMinMaxVarsPerChannel.Options Optional attributes for FakeQuantWithMinMaxVarsPerChannel
FakeQuantWithMinMaxVarsPerChannelGradient Compute gradients for a FakeQuantWithMinMaxVarsPerChannel operation.
FakeQuantWithMinMaxVarsPerChannelGradient.Options Optional attributes for FakeQuantWithMinMaxVarsPerChannelGradient
Quantize <T extends TType > Quantize the 'input' tensor of type float to 'output' tensor of type 'T'.
Quantize.Options Optional attributes for Quantize
QuantizeAndDequantize <T extends TNumber > Quantizes then dequantizes a tensor.
QuantizeAndDequantize.Options Optional attributes for QuantizeAndDequantize
QuantizeAndDequantizeV3 <T extends TNumber > Quantizes then dequantizes a tensor.
QuantizeAndDequantizeV3.Options Optional attributes for QuantizeAndDequantizeV3
QuantizeAndDequantizeV4 <T extends TNumber > Returns the gradient of `quantization.QuantizeAndDequantizeV4`.
QuantizeAndDequantizeV4.Options Optional attributes for QuantizeAndDequantizeV4
QuantizeAndDequantizeV4Grad <T extends TNumber > Returns the gradient of `QuantizeAndDequantizeV4`.
QuantizeAndDequantizeV4Grad.Options Optional attributes for QuantizeAndDequantizeV4Grad
QuantizedConcat <T extends TType > Concatenates quantized tensors along one dimension.
QuantizedMatMulWithBiasAndDequantize <W extends TNumber >
QuantizedMatMulWithBiasAndDequantize.Options Optional attributes for QuantizedMatMulWithBiasAndDequantize
QuantizedMatMulWithBiasAndRequantize <W extends TType >
QuantizedMatMulWithBiasAndRequantize.Options Optional attributes for QuantizedMatMulWithBiasAndRequantize
QuantizeDownAndShrinkRange <U extends TType > Convert the quantized 'input' tensor into a lower-precision 'output', using the

actual distribution of the values to maximize the usage of the lower bit depth and adjusting the output min and max ranges accordingly.

RequantizationRange Computes a range that covers the actual values present in a quantized tensor.
Requantize <U extends TType > Converts the quantized `input` tensor into a lower-precision `output`.