# Copyright 2020-present, Pietro Buzzega, Matteo Boschini, Angelo Porrello, Davide Abati, Simone Calderara.
# All rights reserved.
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
from torch.nn import functional as F
from models.utils.continual_model import ContinualModel
from utils.args import ArgumentParser, add_rehearsal_args
from utils.buffer import Buffer
[docs]
class Der(ContinualModel):
"""Continual learning via Dark Experience Replay."""
NAME = 'der'
COMPATIBILITY = ['class-il', 'domain-il', 'task-il', 'general-continual']
[docs]
@staticmethod
def get_parser(parser) -> ArgumentParser:
add_rehearsal_args(parser)
parser.add_argument('--alpha', type=float, required=True,
help='Penalty weight.')
return parser
def __init__(self, backbone, loss, args, transform, dataset=None):
super(Der, self).__init__(backbone, loss, args, transform, dataset=dataset)
self.buffer = Buffer(self.args.buffer_size)
[docs]
def observe(self, inputs, labels, not_aug_inputs, epoch=None):
self.opt.zero_grad()
tot_loss = 0
outputs = self.net(inputs)
loss = self.loss(outputs, labels)
loss.backward()
tot_loss += loss.item()
if not self.buffer.is_empty():
buf_inputs, buf_logits = self.buffer.get_data(
self.args.minibatch_size, transform=self.transform, device=self.device)
buf_outputs = self.net(buf_inputs)
loss_mse = self.args.alpha * F.mse_loss(buf_outputs, buf_logits)
loss_mse.backward()
tot_loss += loss_mse.item()
self.opt.step()
self.buffer.add_data(examples=not_aug_inputs, logits=outputs.data)
return tot_loss