Learning-Rate-Normalization/adam_lr_norm.py at master · Columbia-Creative-Machines-Lab/Learning-Rate-Normalization · 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
import math
import torch
import numpy as np
from torch.optim import Optimizer


class Adam_lr_norm(Optimizer):
    """Implements Adam algorithm.

    It has been proposed in `Adam: A Method for Stochastic Optimization`_.

    Arguments:
        params (iterable): iterable of parameters to optimize or dicts defining
            parameter groups
        lr (float, optional): learning rate (default: 1e-3)
        betas (Tuple[float, float], optional): coefficients used for computing
            running averages of gradient and its square (default: (0.9, 0.999))
        eps (float, optional): term added to the denominator to improve
            numerical stability (default: 1e-8)
        weight_decay (float, optional): weight decay (L2 penalty) (default: 0)

    .. _Adam\: A Method for Stochastic Optimization:
        https://arxiv.org/abs/1412.6980
    """

    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8,
                 weight_decay=0, schedule=None, gamma=0.1):
        defaults = dict(lr=lr, betas=betas, eps=eps,
                        weight_decay=weight_decay)
        super(Adam_lr_norm, self).__init__(params, defaults)
        self.schedule = schedule
        self.gamma = gamma

    def step(self, closure=None):
        """Performs a single optimization step.

        Arguments:
            closure (callable, optional): A closure that reevaluates the model
                and returns the loss.
        """
        loss = None
        if closure is not None:
            loss = closure()

        d = 0
        weight_norms = []
        bias_norms = []


        for group in self.param_groups:
            #w_mul = np.linalg.norm(self.param_groups[-2].grad.data.cpu().numpy())
            w_mul = 1.0
            b_mul = 1.0

            for p in group['params']:
                if p.grad is None:
                    continue
                grad = p.grad.data

                # normalization
                new_grad = grad
                norm = torch.norm(grad)
                if len(p.data.shape) > 1:
                    if self.schedule == 'exponential':
                        if norm != 0:
                            new_grad = w_mul*new_grad/norm
                            new_grad = new_grad * math.exp(-self.gamma*d)
                    elif self.schedule == 'linear':
                        if norm != 0:
                            new_grad = w_mul*new_grad/norm
                            new_grad = new_grad * ((1-self.gamma)**d)
                    elif self.schedule != 'none':
                        if norm != 0:
                            new_grad = w_mul*new_grad/norm
                    weight_norms.append(norm)
                elif len(p.data.shape) == 1:
                    if self.schedule == 'exponential':
                        if norm != 0:
                            new_grad = b_mul*new_grad/norm
                            new_grad = new_grad * math.exp(-self.gamma*d)
                    elif self.schedule == 'linear':
                        if norm != 0:
                            new_grad = b_mul*new_grad/norm
                            new_grad = new_grad * ((1-self.gamma)**d)
                    elif self.schedule != 'none':
                        if norm != 0:
                            new_grad = b_mul*new_grad/norm
                    bias_norms.append(norm)
                if p.data.shape[0] > 1:
                    d += 1

                grad = new_grad
                if grad.is_sparse:
                    raise RuntimeError('Adam does not support sparse gradients, please consider SparseAdam instead')

                state = self.state[p]

                # State initialization
                if len(state) == 0:
                    state['step'] = 0
                    # Exponential moving average of gradient values
                    state['exp_avg'] = torch.zeros_like(p.data)
                    # Exponential moving average of squared gradient values
                    state['exp_avg_sq'] = torch.zeros_like(p.data)

                exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq']
                beta1, beta2 = group['betas']

                state['step'] += 1

                if group['weight_decay'] != 0:
                    grad = grad.add(group['weight_decay'], p.data)

                # Decay the first and second moment running average coefficient
                exp_avg.mul_(beta1).add_(1 - beta1, grad)
                exp_avg_sq.mul_(beta2).addcmul_(1 - beta2, grad, grad)

                denom = exp_avg_sq.sqrt().add_(group['eps'])

                bias_correction1 = 1 - beta1 ** state['step']
                bias_correction2 = 1 - beta2 ** state['step']
                step_size = group['lr'] * math.sqrt(bias_correction2) / bias_correction1

                p.data.addcdiv_(-step_size, exp_avg, denom)

        return loss, [weight_norms, bias_norms]