elasticai.creator.nn.quantized_grads.fixed_point.param_quantization#

Module Contents#

Classes#

QuantizeTensorToFixedPointHTE

This Modules can be used for Tensor quantization

QuantizeTensorToFixedPointStochastic

This Modules can be used for Tensor quantization

Functions#

get_quantize_to_fixed_point

API#

elasticai.creator.nn.quantized_grads.fixed_point.param_quantization.get_quantize_to_fixed_point(func: Callable[[torch.Tensor, torch.Tensor, torch.Tensor, torch.Tensor], torch.Tensor]) → tuple[type[torch.nn.Module], type[torch.nn.Module]][source]#
class elasticai.creator.nn.quantized_grads.fixed_point.param_quantization.QuantizeTensorToFixedPointHTE(config: elasticai.creator.nn.quantized_grads.fixed_point.two_complement_fixed_point_config.FixedPointConfigV2)[source]#

Bases: elasticai.creator.nn.quantized_grads.fixed_point.param_quantization.QuantizeParamSTEToFixedPointHTE

This Modules can be used for Tensor quantization

Initialization

Initialize internal Module state, shared by both nn.Module and ScriptModule.

class elasticai.creator.nn.quantized_grads.fixed_point.param_quantization.QuantizeTensorToFixedPointStochastic(config: elasticai.creator.nn.quantized_grads.fixed_point.two_complement_fixed_point_config.FixedPointConfigV2)[source]#

Bases: elasticai.creator.nn.quantized_grads.fixed_point.param_quantization.QuantizeParamSTEToFixedPointStochastic

This Modules can be used for Tensor quantization

Initialization

Initialize internal Module state, shared by both nn.Module and ScriptModule.