Commit dd06b323 authored by Gustaf Ahdritz's avatar Gustaf Ahdritz
Browse files

Add barebones EMA

parent ff969b98
from collections import OrderedDict
import copy
import torch
import torch.nn as nn
class ExponentialMovingAverage:
"""
Maintains moving averages of parameters with exponential decay
At each step, the stored copy `copy` of each parameter `param` is
updated as follows:
`copy = decay * copy + (1 - decay) * param`
where `decay` is an attribute of the ExponentialMovingAverage object.
"""
def __init__(self, model: nn.Module, decay: float):
"""
Args:
model:
A torch.nn.Module whose parameters are to be tracked
decay:
A value (usually close to 1.) by which updates are
weighted as part of the above formula
"""
super(ExponentialMovingAverage, self).__init__()
self.params = copy.deepcopy(model.state_dict())
self.decay = decay
def _update_state_dict_(self, update, state_dict):
for k, v in update.items():
stored = state_dict[k]
if(not isinstance(v, torch.Tensor)):
self._update_state_dict_(v, stored)
else:
diff = stored - v
diff *= (1 - self.decay)
stored -= diff
def update(self, model: torch.nn.Module) -> None:
"""
Updates the stored parameters using the state dict of the provided
module. The module should have the same structure as that used to
initialize the ExponentialMovingAverage object.
"""
self._update_state_dict_(model.state_dict(), self.params)
def load_state_dict(self, state_dict: OrderedDict) -> None:
self.params = state_dict["params"]
self.decay = state_dict["decay"]
def state_dict(self) -> OrderedDict:
return OrderedDict({
"params": self.params,
"decay": self.decay,
})
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment