# Source: https://github.com/alphadl/lookahead.pytorch/blob/master/lookahead.py (MIT)
from collections import defaultdict
from torch.optim.optimizer import Optimizer
import torch
[docs]class Lookahead(Optimizer):
r"""
Implementation of `Lookahead Optimizer: k steps forward, 1 step back <https://arxiv.org/abs/1907.08610>`_
Args:
:param optimizer: - the optimizer to work with (sgd, adam etc)
:param k: (int) - number of steps to look ahead (default=5)
:param alpha: (float) - slow weights step size
"""
def __init__(self, optimizer, k=5, alpha=0.5):
"""
:param optimizer: - the optimizer to work with (sgd, adam etc)
:param k: (int) - number of steps to look ahead (default=5)
:param alpha: (float) - slow weights step size
"""
self.optimizer = optimizer
self.k = k
self.alpha = alpha
self.param_groups = self.optimizer.param_groups
self.state = defaultdict(dict)
self.fast_state = self.optimizer.state
for group in self.param_groups:
group["counter"] = 0
[docs] def update(self, group):
for fast in group["params"]:
param_state = self.state[fast]
if "slow_param" not in param_state:
param_state["slow_param"] = torch.zeros_like(fast.data)
param_state["slow_param"].copy_(fast.data)
slow = param_state["slow_param"]
slow += (fast.data - slow) * self.alpha
fast.data.copy_(slow)
[docs] def update_lookahead(self):
for group in self.param_groups:
self.update(group)
[docs] def step(self, closure=None):
loss = self.optimizer.step(closure)
for group in self.param_groups:
if group["counter"] == 0:
self.update(group)
group["counter"] += 1
if group["counter"] >= self.k:
group["counter"] = 0
return loss
[docs] def state_dict(self):
fast_state_dict = self.optimizer.state_dict()
slow_state = {
(id(k) if isinstance(k, torch.Tensor) else k): v
for k, v in self.state.items()
}
fast_state = fast_state_dict["state"]
param_groups = fast_state_dict["param_groups"]
return {
"fast_state": fast_state,
"slow_state": slow_state,
"param_groups": param_groups,
}
[docs] def load_state_dict(self, state_dict):
slow_state_dict = {
"state": state_dict["slow_state"],
"param_groups": state_dict["param_groups"],
}
fast_state_dict = {
"state": state_dict["fast_state"],
"param_groups": state_dict["param_groups"],
}
super(Lookahead, self).load_state_dict(slow_state_dict)
self.optimizer.load_state_dict(fast_state_dict)
self.fast_state = self.optimizer.state
[docs] def add_param_group(self, param_group):
param_group["counter"] = 0
self.optimizer.add_param_group(param_group)