utils.py

import torch
import torch.utils.data as data
import torch.nn.functional as F
import numpy as np

try:
    from itertools import ifilterfalse
except ImportError:  # py3k
    from itertools import filterfalse as ifilterfalse


def overall_acc(pred, label):   # overall accuracy
    r = (torch.argmax(pred, dim=-1) == label).float()
    acc = torch.mean(r)
    return acc

class AvgMeter(object):
    def __init__(self, num=40):
        self.num = num
        self.val = 0
        self.avg = 0
        self.sum = 0
        self.count = 0
        self.losses = []

    def reset(self):
        self.val = 0
        self.avg = 0
        self.sum = 0
        self.count = 0
        self.losses = []

    def update(self, val, n=1):
        self.val = val
        self.sum += val * n
        self.count += n
        self.avg = self.sum / self.count
        self.losses.append(val)

    def show(self):
        return torch.mean(torch.stack(self.losses[np.maximum(len(self.losses) - self.num, 0):]))