mltk.core.TfliteFullyConnectedParams

class TfliteFullyConnectedParams[source]

Calculated Full Connected Parameters

Properties

input_offset

Input quantization offset (i.e.

output_multiplier

Output multiplier for its quantization scaler

output_offset

Output quantization offset (i.e.

output_shift

Output shift for its quantization scaler

quantized_activation_max

Fused activation max value

quantized_activation_min

Fused activation min value

weights_offset

Weight (aka filters) quantization offset (i.e.

Methods

__init__

calculate

Calculate the parameters for the given layer

__init__(input_offset=0, weights_offset=0, output_offset=0, output_multiplier=0, output_shift=0, quantized_activation_min=0, quantized_activation_max=0)
Parameters:
  • input_offset (int) –

  • weights_offset (int) –

  • output_offset (int) –

  • output_multiplier (int) –

  • output_shift (int) –

  • quantized_activation_min (int) –

  • quantized_activation_max (int) –

Return type:

None

input_offset: int = 0

Input quantization offset (i.e. zero point)

weights_offset: int = 0

Weight (aka filters) quantization offset (i.e. zero point)

output_offset: int = 0

Output quantization offset (i.e. zero point)

output_multiplier: int = 0

Output multiplier for its quantization scaler

output_shift: int = 0

Output shift for its quantization scaler

quantized_activation_min: int = 0

Fused activation min value

quantized_activation_max: int = 0

Fused activation max value

static calculate(layer)[source]

Calculate the parameters for the given layer

Return type:

TfliteFullyConnectedParams