Rate this Page

float16_dynamic_qconfig#

torch.ao.quantization.qconfig.float16_dynamic_qconfig#

Dynamic qconfig with weights quantized to torch.float16.