determine_qparams

class torch.ao.quantization.utils.determine_qparams(min_val, max_val, quant_min, quant_max, dtype, eps, has_customized_qrange, qscheme=torch.per_tensor_affine)

Calculates the quantization parameters, given min and max value tensors. Works for both per tensor and per channel cases

Parameters

• min_val (Tensor) – Minimum values per channel

• max_val (Tensor) – Maximum values per channel

Returns

Scales tensor of shape (#channels,) zero_points: Zero points tensor of shape (#channels,)

Return type

scales