iicd/xautodl
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Prototype MAML

Browse Source
This commit is contained in:
D-X-Y 2021-05-10 09:42:42 +08:00
parent 5457dcf042
commit 755c7c90cf
1 changed files with 67 additions and 75 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

142
exps/LFNA/basic-maml.py
View File

@ -1,8 +1,8 @@
#####################################################
# Copyright (c) Xuanyi Dong [GitHub D-X-Y], 2021.04 #
#####################################################
# python exps/LFNA/basic-maml.py --env_version v1 #
# python exps/LFNA/basic-maml.py --env_version v2 #
# python exps/LFNA/basic-maml.py --env_version v1 --hidden_dim 16 --inner_step 5
# python exps/LFNA/basic-maml.py --env_version v2
#####################################################
import sys, time, copy, torch, random, argparse
from tqdm import tqdm
@ -30,73 +30,71 @@ from lfna_utils import lfna_setup, TimeData
class MAML:
"""A LFNA meta-model that uses the MLP as delta-net."""
def __init__(self, container, criterion, meta_lr, inner_lr=0.01, inner_step=1):
def __init__(
self, network, criterion, epochs, meta_lr, inner_lr=0.01, inner_step=1
):
self.criterion = criterion
self.container = container
# self.container = container
self.network = network
self.meta_optimizer = torch.optim.Adam(
self.container.parameters(), lr=meta_lr, amsgrad=True
self.network.parameters(), lr=meta_lr, amsgrad=True
)
self.meta_lr_scheduler = torch.optim.lr_scheduler.MultiStepLR(
optimizer,
milestones=[
int(epochs * 0.25),
int(epochs * 0.5),
int(epochs * 0.75),
],
gamma=0.3,
)
self.inner_lr = inner_lr
self.inner_step = inner_step
self._best_info = dict(state_dict=None, score=None)
print("There are {:} weights.".format(w_container.numel()))
def adapt(self, model, dataset):
def adapt(self, dataset):
# create a container for the future timestamp
y_hat = model.forward_with_container(dataset.x, self.container)
loss = self.criterion(y_hat, dataset.y)
grads = torch.autograd.grad(loss, self.container.parameters())
container = self.network.get_w_container()
fast_container = self.container.additive(
[-self.inner_lr * grad for grad in grads]
)
import pdb
for k in range(0, self.inner_step):
y_hat = self.network.forward_with_container(dataset.x, container)
loss = self.criterion(y_hat, dataset.y)
grads = torch.autograd.grad(loss, container.parameters())
pdb.set_trace()
w_container.requires_grad_(True)
containers = [w_container]
for idx, dataset in enumerate(seq_datasets):
x, y = dataset.x, dataset.y
y_hat = model.forward_with_container(x, containers[-1])
loss = criterion(y_hat, y)
gradients = torch.autograd.grad(loss, containers[-1].tensors)
with torch.no_grad():
flatten_w = containers[-1].flatten().view(-1, 1)
flatten_g = containers[-1].flatten(gradients).view(-1, 1)
input_statistics = torch.tensor([x.mean(), x.std()]).view(1, 2)
input_statistics = input_statistics.expand(flatten_w.numel(), -1)
delta_inputs = torch.cat((flatten_w, flatten_g, input_statistics), dim=-1)
delta = self.delta_net(delta_inputs).view(-1)
delta = torch.clamp(delta, -0.5, 0.5)
unflatten_delta = containers[-1].unflatten(delta)
future_container = containers[-1].no_grad_clone().additive(unflatten_delta)
# future_container = containers[-1].additive(unflatten_delta)
containers.append(future_container)
# containers = containers[1:]
meta_loss = []
temp_containers = []
for idx, dataset in enumerate(seq_datasets):
if idx == 0:
continue
current_container = containers[idx]
y_hat = model.forward_with_container(dataset.x, current_container)
loss = criterion(y_hat, dataset.y)
meta_loss.append(loss)
temp_containers.append((dataset.timestamp, current_container, -loss.item()))
meta_loss = sum(meta_loss)
w_container.requires_grad_(False)
# meta_loss.backward()
# self.meta_optimizer.step()
return meta_loss, temp_containers
container = container.additive([-self.inner_lr * grad for grad in grads])
return container
def predict(self, x, container=None):
if container is not None:
y_hat = self.network.forward_with_container(x, container)
else:
y_hat = self.network(x)
return y_hat
def step(self):
torch.nn.utils.clip_grad_norm_(self.delta_net.parameters(), 1.0)
torch.nn.utils.clip_grad_norm_(self.container.parameters(), 1.0)
self.meta_optimizer.step()
self.meta_lr_scheduler.step()
def zero_grad(self):
self.meta_optimizer.zero_grad()
def save_best(self, network, score):
if self._best_info["score"] is None or self._best_info["score"] < score:
state_dict = dict(
criterion=criterion,
network=network.state_dict(),
meta_optimizer=self.meta_optimizer.state_dict(),
meta_lr_scheduler=self.meta_lr_scheduler.state_dict(),
)
self._best_info["state_dict"] = state_dict
self._best_info["score"] = score
def main(args):
logger, env_info = lfna_setup(args)
logger, env_info, model_kwargs = lfna_setup(args)
model = get_model(dict(model_type="simple_mlp"), **model_kwargs)
total_time = env_info["total"]
for i in range(total_time):
@ -104,19 +102,12 @@ def main(args):
nkey = "{:}-{:}".format(i, xkey)
assert nkey in env_info, "{:} no in {:}".format(nkey, list(env_info.keys()))
train_time_bar = total_time // 2
base_model = get_model(
dict(model_type="simple_mlp"),
act_cls="leaky_relu",
norm_cls="identity",
input_dim=1,
output_dim=1,
)
w_container = base_model.get_w_container()
criterion = torch.nn.MSELoss()
print("There are {:} weights.".format(w_container.numel()))
maml = MAML(w_container, criterion, args.meta_lr, args.inner_lr, args.inner_step)
maml = MAML(
model, criterion, args.epochs, args.meta_lr, args.inner_lr, args.inner_step
)
# meta-training
per_epoch_time, start_time = AverageMeter(), time.time()
@ -131,8 +122,7 @@ def main(args):
)
maml.zero_grad()
all_meta_losses = []
meta_losses = []
for ibatch in range(args.meta_batch):
sampled_timestamp = random.randint(0, train_time_bar)
past_dataset = TimeData(
@ -145,21 +135,23 @@ def main(args):
env_info["{:}-x".format(sampled_timestamp + 1)],
env_info["{:}-y".format(sampled_timestamp + 1)],
)
maml.adapt(base_model, past_dataset)
import pdb
pdb.set_trace()
meta_loss = torch.stack(all_meta_losses).mean()
future_container = maml.adapt(model, past_dataset)
future_y_hat = maml.predict(future_dataset.x, future_container)
future_loss = maml.criterion(future_y_hat, future_dataset.y)
meta_losses.append(future_loss)
meta_loss = torch.stack(meta_losses).mean()
meta_loss.backward()
adaptor.step()
maml.step()
debug_str = pool.debug_info(debug_timestamp)
logger.log("meta-loss: {:.4f}".format(meta_loss.item()))
per_epoch_time.update(time.time() - start_time)
start_time = time.time()
import pdb
pdb.set_trace()
logger.log("-" * 200 + "\n")
logger.close()
@ -187,7 +179,7 @@ if __name__ == "__main__":
parser.add_argument(
"--meta_lr",
type=float,
default=0.01,
default=0.1,
help="The learning rate for the MAML optimizer (default is Adam)",
)
parser.add_argument(
@ -202,7 +194,7 @@ if __name__ == "__main__":
parser.add_argument(
"--meta_batch",
type=int,
default=5,
default=10,
help="The batch size for the meta-model",
)
parser.add_argument(
@ -223,7 +215,7 @@ if __name__ == "__main__":
if args.rand_seed is None or args.rand_seed < 0:
args.rand_seed = random.randint(1, 100000)
assert args.save_dir is not None, "The save dir argument can not be None"
args.save_dir = "{:}-{:}-d{:}".format(
args.save_dir, args.env_version, args.hidden_dim
args.save_dir = "{:}-s{:}-{:}-d{:}".format(
args.save_dir, args.inner_step, args.env_version, args.hidden_dim
)
main(args)