from utils.buffer import Buffer
from torch.nn import functional as F
from utils.args import *
import torch
from models.utils.lider_model import LiderOptimizer, add_lipschitz_args
[docs]
class DerppLider(LiderOptimizer):
"""Continual learning via Dark Experience Replay++. Treated with LiDER!"""
NAME = 'derpp_lider'
COMPATIBILITY = ['class-il', 'domain-il', 'task-il', 'general-continual']
[docs]
@staticmethod
def get_parser(parser) -> ArgumentParser:
add_rehearsal_args(parser)
add_lipschitz_args(parser)
parser.add_argument('--alpha', type=float, required=True,
help='Penalty weight.')
parser.add_argument('--beta', type=float, required=True,
help='Penalty weight.')
return parser
def __init__(self, backbone, loss, args, transform, dataset=None):
super().__init__(backbone, loss, args, transform, dataset=dataset)
self.buffer = Buffer(self.args.buffer_size)
[docs]
def begin_task(self, dataset):
if self.current_task == 0:
self.net.set_return_prerelu(True)
self.init_net(dataset)
[docs]
def observe(self, inputs: torch.Tensor, labels: torch.Tensor, not_aug_inputs: torch.Tensor, epoch=None):
self.opt.zero_grad()
outputs = self.net(inputs)
loss = self.loss(outputs, labels)
tot_loss = loss.item()
loss.backward()
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, buf_output_features = self.net(buf_inputs, returnt='full')
loss_mse = self.args.alpha * F.mse_loss(buf_outputs, buf_logits)
tot_loss += loss_mse.item()
loss_mse.backward()
buf_inputs, buf_labels, _ = self.buffer.get_data(
self.args.minibatch_size, transform=self.transform, device=self.device)
buf_outputs = self.net(buf_inputs).float()
loss_ce = self.args.beta * self.loss(buf_outputs, buf_labels)
tot_loss += loss_ce.item()
loss_ce.backward()
if self.args.alpha_lip_lambda > 0:
buf_inputs, _, _ = self.buffer.get_data(self.args.minibatch_size, transform=self.transform, device=self.device)
_, buf_output_features = self.net(buf_inputs, returnt='full')
lip_inputs = [buf_inputs] + buf_output_features
loss_lip_minimize = self.args.alpha_lip_lambda * self.minimization_lip_loss(lip_inputs)
tot_loss += loss_lip_minimize.item()
loss_lip_minimize.backward()
if self.args.beta_lip_lambda > 0:
buf_inputs, _, _ = self.buffer.get_data(self.args.minibatch_size, transform=self.transform, device=self.device)
_, buf_output_features = self.net(buf_inputs, returnt='full')
lip_inputs = [buf_inputs] + buf_output_features
loss_lip_dyn_budget = self.args.beta_lip_lambda * self.dynamic_budget_lip_loss(lip_inputs)
tot_loss += loss_lip_dyn_budget.item()
loss_lip_dyn_budget.backward()
self.opt.step()
self.buffer.add_data(examples=not_aug_inputs,
labels=labels,
logits=outputs.data)
return tot_loss