# Copyright (c) 2020 Horizon Robotics and ALF Contributors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import math
import torch
from torch.optim import Optimizer
from .utils import get_opt_arg
[docs]class AdamTF(Optimizer):
r"""Implementation of Adam algorithm following Tensorflow's convention.
This class should not be direclty used as it will be wrapped for clipping
gradients. Use the wrapped optimizer ``AdamTF`` in ``alf/optimizers/optimizers.py``
instead.
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 which corresponds to the
epsilon_hat in the Adam paper (default: 1e-7).
weight_decay (float, optional): weight decay (L2 penalty) (default: 0).
This argument can be parameter specific, which means that if
Parameter.opt_args["weight_decay"] is not None, it will be used instead.
amsgrad (boolean, optional): whether to use the AMSGrad variant of this
algorithm from the paper `On the Convergence of Adam and Beyond`_
(default: False).
References:
.. _Adam\: A Method for Stochastic Optimization:
https://arxiv.org/abs/1412.6980
.. _On the Convergence of Adam and Beyond:
https://openreview.net/forum?id=ryQu7f-RZ
"""
def __init__(self,
params=[{
'params': []
}],
lr=1e-3,
betas=(0.9, 0.999),
eps=1e-7,
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().__init__(params, defaults)
self._state_ready = False
def __setstate__(self, state):
super().__setstate__(state)
for group in self.param_groups:
group.setdefault('amsgrad', False)
[docs] def reset_state(self):
"""Performs reset to all the states of the AdamTF optimizer, including
exponential moving average of gradients and squared gradients etc.
"""
for group in self.param_groups:
amsgrad = group['amsgrad']
for p in group['params']:
state = self.state[p]
state['step'] = 0
# Exponential moving average of gradient values
state['exp_avg'] = torch.zeros_like(
p, memory_format=torch.preserve_format)
# Exponential moving average of squared gradient values
state['exp_avg_sq'] = torch.zeros_like(
p, memory_format=torch.preserve_format)
if 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)
[docs] @torch.no_grad()
def step(self, closure=None):
"""Performs a single optimization step.
Arguments:
closure (callable, optional): A closure that reevaluates the model
and returns the loss.
"""
if not self._state_ready:
self.reset_state()
self._state_ready = True
loss = None
if closure is not None:
with torch.enable_grad():
loss = closure()
for group in self.param_groups:
for p in group['params']:
if p.grad is None:
continue
grad = p.grad
if grad.is_sparse:
raise RuntimeError(
'Adam does not support sparse gradients, please consider SparseAdam instead'
)
amsgrad = group['amsgrad']
state = self.state[p]
exp_avg, exp_avg_sq = state['exp_avg'], state['exp_avg_sq']
if amsgrad:
max_exp_avg_sq = state['max_exp_avg_sq']
beta1, beta2 = group['betas']
state['step'] += 1
bias_correction1 = 1 - beta1**state['step']
bias_correction2 = 1 - beta2**state['step']
weight_decay = get_opt_arg(p, 'weight_decay',
group['weight_decay'])
if weight_decay != 0:
grad = grad.add(p, alpha=weight_decay)
# Decay the first and second moment running average coefficient
exp_avg.lerp_(grad, 1 - beta1)
exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
if amsgrad:
# Maintains the maximum of all 2nd moment running avg. till now
torch.max(max_exp_avg_sq, exp_avg_sq, out=max_exp_avg_sq)
# Use the max. for normalizing running avg. of gradient
denom = max_exp_avg_sq.sqrt().add_(group['eps'])
else:
denom = exp_avg_sq.sqrt().add_(group['eps'])
step_size = group['lr'] * (
math.sqrt(bias_correction2) / bias_correction1)
p.addcdiv_(exp_avg, denom, value=-step_size)
return loss