quantize_qat¶
-
class
torch.quantization.quantize_qat(model, run_fn, run_args, inplace=False)[source]¶ Do quantization aware training and output a quantized model
- Parameters
model – input model
run_fn – a function for evaluating the prepared model, can be a function that simply runs the prepared model or a training loop
run_args – positional arguments for run_fn
- Returns
Quantized model.
