optim.py

import math

import torch
from torch.optim import Optimizer
from torch import Tensor
from typing import List


def adam(params: List[Tensor],
         grads: List[Tensor],
         exp_avgs: List[Tensor],
         exp_avg_sqs: List[Tensor],
         max_exp_avg_sqs: List[Tensor],
         state_steps: List[int],
         amsgrad: bool,
         beta1: float,
         beta2: float,
         lr: float,
         weight_decay: float,
         eps: float):
    r"""Functional API that performs Adam algorithm computation.
    See :class:`~torch.optim.Adam` for details.
    """

    for i, param in enumerate(params):

        grad = grads[i]
        if beta1 != 0:
            exp_avg = exp_avgs[i]
        if beta2 != 0:
            exp_avg_sq = exp_avg_sqs[i]
        step = state_steps[i]

        bias_correction1 = 1 - beta1 ** step
        bias_correction2 = 1 - beta2 ** step

        if weight_decay != 0:
            grad = grad.add(param, alpha=weight_decay)

        # Decay the first and second moment running average coefficient
        if beta1 != 0:
            exp_avg.mul_(beta1).add_(grad, alpha=1 - beta1)
        else:
            exp_avg = grad
        if beta2 != 0:
            exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
        else:
            exp_avg_sq = grad * grad
        if amsgrad:
            # Maintains the maximum of all 2nd moment running avg. till now
            torch.maximum(
                max_exp_avg_sqs[i], exp_avg_sq, out=max_exp_avg_sqs[i])
            # Use the max. for normalizing running avg. of gradient
            denom = (
                max_exp_avg_sqs[i].sqrt() / math.sqrt(bias_correction2)
            ).add_(eps)
        else:
            denom = (exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)

        step_size = lr / bias_correction1
        param.addcdiv_(exp_avg, denom, value=-step_size)


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

    It has been proposed in `Adam: A Method for Stochastic Optimization`_.
    The implementation of the L2 penalty follows changes proposed in
    `Decoupled Weight Decay Regularization`_.

    Args:
        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)
        amsgrad (boolean, optional): whether to use the AMSGrad variant of this
            algorithm from the paper `On the Convergence of Adam and Beyond`_
            (default: False)

    .. _Adam\: A Method for Stochastic Optimization:
        https://arxiv.org/abs/1412.6980
    .. _Decoupled Weight Decay Regularization:
        https://arxiv.org/abs/1711.05101
    .. _On the Convergence of Adam and Beyond:
        https://openreview.net/forum?id=ryQu7f-RZ
    """

    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8,
                 weight_decay=0, amsgrad=False):
        if not 0.0 <= lr:
            raise ValueError("Invalid learning rate: {}".format(lr))
        if not 0.0 <= eps:
            raise ValueError("Invalid epsilon value: {}".format(eps))
        if not 0.0 <= betas[0] < 1.0:
            raise ValueError(
                "Invalid beta parameter at index 0: {}".format(betas[0]))
        if not 0.0 <= betas[1] < 1.0:
            raise ValueError(
                "Invalid beta parameter at index 1: {}".format(betas[1]))
        if not 0.0 <= weight_decay:
            raise ValueError(
                "Invalid weight_decay value: {}".format(weight_decay))
        defaults = dict(lr=lr, betas=betas, eps=eps,
                        weight_decay=weight_decay, amsgrad=amsgrad)
        super(Adam, self).__init__(params, defaults)

    def __setstate__(self, state):
        super(Adam, self).__setstate__(state)
        for group in self.param_groups:
            group.setdefault('amsgrad', False)

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

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

        for group in self.param_groups:
            beta1, beta2 = group['betas']
            params_with_grad = []
            grads = []
            if beta1 != 0:
                exp_avgs = []
            else:
                exp_avgs = None
            if beta2 != 0:
                exp_avg_sqs = []
            else:
                exp_avg_sqs = None
            max_exp_avg_sqs = []
            state_steps = []

            for p in group['params']:
                if p.grad is not None:
                    params_with_grad.append(p)
                    if p.grad.is_sparse:
                        raise RuntimeError('Adam does not support sparse gradients, please consider SparseAdam instead')
                    grads.append(p.grad)

                    state = self.state[p]
                    # Lazy state initialization
                    if len(state) == 0:
                        state['step'] = 0
                        # Exponential moving average of gradient values
                        if beta1 != 0:
                            state['exp_avg'] = torch.zeros_like(
                                p, memory_format=torch.preserve_format)
                        # Exponential moving average of squared gradient values
                        if beta2 != 0:
                            state['exp_avg_sq'] = torch.zeros_like(
                                p, memory_format=torch.preserve_format)
                        if group['amsgrad']:
                            # Maintains max of all exp. moving avg. of sq.
                            # grad. values
                            state['max_exp_avg_sq'] = torch.zeros_like(
                                p, memory_format=torch.preserve_format)
                    if beta1 != 0:
                        exp_avgs.append(state['exp_avg'])
                    if beta2 != 0:
                        exp_avg_sqs.append(state['exp_avg_sq'])

                    if group['amsgrad']:
                        max_exp_avg_sqs.append(state['max_exp_avg_sq'])

                    # update the steps for each param group update
                    state['step'] += 1
                    # record the step after step update
                    state_steps.append(state['step'])

            adam(params_with_grad,
                 grads,
                 exp_avgs,
                 exp_avg_sqs,
                 max_exp_avg_sqs,
                 state_steps,
                 group['amsgrad'],
                 beta1,
                 beta2,
                 group['lr'],
                 group['weight_decay'],
                 group['eps'])
        return loss