Source code for xuance.torch.utils.value_norm

import numpy as np
import torch
import torch.nn as nn




class ValueNorm(nn.Module):
    """ Normalize a vector of observations - across the first norm_axes dimensions"""

    def __init__(self, input_shape, norm_axes=1, beta=0.99999, per_element_update=False, epsilon=1e-5):
        super(ValueNorm, self).__init__()

        self.input_shape = input_shape
        self.norm_axes = norm_axes
        self.epsilon = epsilon
        self.beta = beta
        self.per_element_update = per_element_update

        self.running_mean = nn.Parameter(torch.zeros(input_shape), requires_grad=False)
        self.running_mean_sq = nn.Parameter(torch.zeros(input_shape), requires_grad=False)
        self.debiasing_term = nn.Parameter(torch.tensor(0.0), requires_grad=False)

        self.reset_parameters()



    def reset_parameters(self):
        self.running_mean.zero_()
        self.running_mean_sq.zero_()
        self.debiasing_term.zero_()




    def running_mean_var(self):
        debiased_mean = self.running_mean / self.debiasing_term.clamp(min=self.epsilon)
        debiased_mean_sq = self.running_mean_sq / self.debiasing_term.clamp(min=self.epsilon)
        debiased_var = (debiased_mean_sq - debiased_mean ** 2).clamp(min=1e-2)
        return debiased_mean, debiased_var


    @torch.no_grad()
    def update(self, input_vector):
        if type(input_vector) == np.ndarray:
            input_vector = torch.from_numpy(input_vector)
        input_vector = input_vector.to(self.running_mean.device)  # not elegant, but works in most cases

        batch_mean = input_vector.mean(dim=tuple(range(self.norm_axes)))
        batch_sq_mean = (input_vector ** 2).mean(dim=tuple(range(self.norm_axes)))

        if self.per_element_update:
            batch_size = np.prod(input_vector.size()[:self.norm_axes])
            weight = self.beta ** batch_size
        else:
            weight = self.beta

        self.running_mean.mul_(weight).add_(batch_mean * (1.0 - weight))
        self.running_mean_sq.mul_(weight).add_(batch_sq_mean * (1.0 - weight))
        self.debiasing_term.mul_(weight).add_(1.0 * (1.0 - weight))



    def normalize(self, input_vector):
        # Make sure input is float32
        if type(input_vector) == np.ndarray:
            input_vector = torch.from_numpy(input_vector)
        input_vector = input_vector.to(self.running_mean.device)  # not elegant, but works in most cases

        mean, var = self.running_mean_var()
        out = (input_vector - mean[(None,) * self.norm_axes]) / torch.sqrt(var)[(None,) * self.norm_axes]

        return out




    def denormalize(self, input_vector):
        """ Transform normalized data back into original distribution """
        input_vector = torch.as_tensor(input_vector)
        input_vector = input_vector.to(self.running_mean.device)  # not elegant, but works in most cases

        mean, var = self.running_mean_var()
        out = input_vector * torch.sqrt(var)[(None,) * self.norm_axes] + mean[(None,) * self.norm_axes]

        out = out.cpu().numpy()

        return out