MCPcopy Create free account
hub / github.com/pytorch/executorch / QuantizationConfig

Class QuantizationConfig

backends/samsung/quantizer/qconfig.py:26–30  ·  view source on GitHub ↗

Source from the content-addressed store, hash-verified

24
25@dataclass(eq=True, frozen=True)
26class QuantizationConfig:
27 input_activation: Optional[QuantizationSpec]
28 output_activation: Optional[QuantizationSpec]
29 weight: Optional[QuantizationSpec]
30 bias: Optional[QuantizationSpec | Callable]
31
32
33def get_quant_config(

Callers 8

__init__Method · 0.90
test_per_channel_convolutionMethod · 0.90
get_symmetric_quantization_configFunction · 0.90
get_symmetric_quantization_configFunction · 0.90
quantizer.pyFile · 0.90
get_custom_quantizerFunction · 0.90
add_mse_weight_observerFunction · 0.90
get_a8w8_enn_quant_configFunction · 0.70

Calls

no outgoing calls

Tested by 1

test_per_channel_convolutionMethod · 0.72