videobagnet/utils.py at main · ombretta/videobagnet · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
import csv
import random
from functools import partialmethod

import torch
from torchmetrics.classification import Accuracy, AveragePrecision
#from torchmetrics.functional import Accuracy, AveragePrecision
import numpy as np
from sklearn.metrics import precision_recall_fscore_support


def count_parameters(model):
    return sum(p.numel() for p in model.parameters() if p.requires_grad)

def calculate_RF_size(model):
    named_layers = dict(model.named_modules())
    downsampling = []
    l = []  # [[1, 1, 1]]
    i = 0

    for layer in named_layers.keys():
        curr_layer = named_layers[layer]
        # n.b.:
        if 'kernel_size' in curr_layer.__dict__ and 'downsample' not in layer:
            k = curr_layer.kernel_size
            s = curr_layer.stride
            if type(k) == int: k, s = (k, k, k), (s, s, s)
            downsampling.append({'f': k, 's': s})

            dims = len(downsampling[i]['f'])
            if l == []: l = [[1] * dims]

            si, li = [], []
            for dim in range(min(dims, len(l[i]))):
                si_d = np.prod([x['s'][dim] for x in downsampling[:i]])
                li_d = l[i][dim] + (downsampling[i]['f'][dim] - 1) * si_d
                si += [si_d]
                li += [li_d]
            l.append(li)
            print("%d) The receptive field after %s is %s"
                  % (i + 1, layer, li))
            i += 1


class AverageMeter(object):
    """Computes and stores the average and current value"""

    def __init__(self):
        self.reset()

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

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


class Logger(object):

    def __init__(self, path, header):
        self.log_file = path.open('w')
        self.logger = csv.writer(self.log_file, delimiter='\t')

        self.logger.writerow(header)
        self.header = header

    def __del(self):
        self.log_file.close()

    def log(self, values):
        write_values = []
        for col in self.header:
            assert col in values
            write_values.append(values[col])

        self.logger.writerow(write_values)
        self.log_file.flush()


def calculate_accuracy(outputs, targets, multilabel=False):
    with torch.no_grad():
        batch_size = targets.size(0)

        _, pred = outputs.topk(1, 1, largest=True, sorted=True)
        #print("outputs", outputs, "pred", pred)
        pred = pred.t()
        correct = pred.eq(targets.view(1, -1))
        n_correct_elems = correct.float().sum().item()

        return n_correct_elems / batch_size

def calculate_accuracy_pytorch(outputs, targets, multilabel=False, n_classes=10):
    with torch.no_grad():
        if not multilabel:
            pred = torch.nn.Softmax(dim=1)(outputs).cpu()
            #print(pred, targets)
            accuracy = Accuracy(task="multiclass", num_classes=n_classes)
            return accuracy(pred, targets.to('cpu', dtype=torch.int))
        else:
            pred = torch.nn.Sigmoid()(outputs).cpu()
            #print(pred, targets.to('cpu', dtype=torch.int))
            average_precision = AveragePrecision(task="multiclass", num_classes=n_classes, average='macro')
            AP = average_precision(pred, targets.to('cpu', dtype=torch.int))
            return AP

def calculate_precision_and_recall(outputs, targets, pos_label=1):
    with torch.no_grad():
        _, pred = outputs.topk(1, 1, largest=True, sorted=True)
        precision, recall, _, _ = precision_recall_fscore_support(
            targets.view(-1, 1).cpu().numpy(),
            pred.cpu().numpy())

        return precision[pos_label], recall[pos_label]


def worker_init_fn(worker_id):
    torch_seed = torch.initial_seed()

    random.seed(torch_seed + worker_id)

    if torch_seed >= 2**32:
        torch_seed = torch_seed % 2**32
    np.random.seed(torch_seed + worker_id)


def get_lr(optimizer):
    lrs = []
    for param_group in optimizer.param_groups:
        lr = float(param_group['lr'])
        lrs.append(lr)

    return max(lrs)


def partialclass(cls, *args, **kwargs):

    class PartialClass(cls):
        __init__ = partialmethod(cls.__init__, *args, **kwargs)

    return PartialClass