def testing_phase(self) -> None:
"""
Evaluates model using test set data.
"""
self.test_dataloader = self.get_dataloader(self.test_h5_path,
"test set")
self.load_training_set_properties()
self.restart_epoch = util.get_restart_epoch()
print(
f"* Loading model from previous saved state (Epoch {self.restart_epoch}).",
flush=True)
model_path = self.constants.job_dir + f"model_restart_{self.restart_epoch}.pth"
self.model = self.create_model()
try:
# for loading models created using GraphINVENT v1.0 (will raise an exception
# if model was created with GraphINVENT v2.0)
self.model.state_dict = torch.load(model_path).state_dict()
except AttributeError:
# for loading models created using GraphINVENT v2.0
self.model.load_state_dict(torch.load(model_path))
self.model.eval()
with torch.no_grad():
self.generate_graphs(n_samples=self.constants.n_samples)
print("* Evaluating model.", flush=True)
self.analyzer.model = self.model
self.analyzer.evaluate_model(nll_per_action=self.nll_per_action)
self.print_time_elapsed()
def testing_phase(self):
""" Evaluates model using test set data.
"""
self.test_dataloader = self.get_dataloader(self.test_h5_path,
"test set")
self.get_ts_properties()
self.restart_epoch = util.get_restart_epoch()
print(
f"* Loading model from previous saved state (Epoch {self.restart_epoch}).",
flush=True)
self.model = torch.load(self.C.job_dir +
f"model_restart_{self.restart_epoch}.pth")
self.model.eval()
with torch.no_grad():
self.generate_graphs(n_samples=self.C.n_samples)
print("* Evaluating model.", flush=True)
anal.evaluate_model(valid_dataloader=self.test_dataloader,
train_dataloader=self.train_dataloader,
nll_per_action=self.nll_per_action,
model=self.model)
self.print_time_elapsed()
def define_model_and_optimizer(self) -> Tuple[time.time, time.time]:
"""
Defines the model, optimizer, and scheduler.
"""
print("* Defining model.", flush=True)
job_dir = self.constants.job_dir
self.model = self.create_model()
if self.constants.restart:
print("-- Loading model from previous saved state.", flush=True)
self.restart_epoch = util.get_restart_epoch()
try:
# for loading models created using GraphINVENT v1.0 (will raise an exception
# if model was created with GraphINVENT v2.0)
self.model.state_dict = torch.load(
f"{job_dir}model_restart_{self.restart_epoch}.pth"
).state_dict()
except AttributeError:
# for loading models created using GraphINVENT v2.0
self.model.load_state_dict(
torch.load(
f"{job_dir}model_restart_{self.restart_epoch}.pth"))
print(
f"-- Backing up as {job_dir}model_restart_{self.restart_epoch}_restarted.pth.",
flush=True)
shutil.copyfile(
f"{job_dir}model_restart_{self.restart_epoch}.pth",
f"{job_dir}model_restart_{self.restart_epoch}_restarted.pth",
)
else:
self.restart_epoch = 0
start_epoch = self.restart_epoch + 1
end_epoch = start_epoch + self.constants.epochs
print("-- Defining optimizer.", flush=True)
self.optimizer = torch.optim.Adam(params=self.model.parameters(),
lr=self.constants.init_lr)
print("-- Defining scheduler.", flush=True)
max_allowable_lr = self.constants.max_rel_lr * self.constants.init_lr
n_batches = len(self.train_dataloader)
self.scheduler = torch.optim.lr_scheduler.OneCycleLR(
optimizer=self.optimizer,
max_lr=max_allowable_lr,
steps_per_epoch=n_batches,
epochs=self.constants.epochs)
return start_epoch, end_epoch
def define_model_and_optimizer(self):
""" Defines the model (`self.model`) and the optimizer (`self.optimizer`).
"""
print("* Defining model and optimizer.", flush=True)
job_dir = self.C.job_dir
if self.C.restart:
print("-- Loading model from previous saved state.", flush=True)
self.restart_epoch = util.get_restart_epoch()
self.model = torch.load(
f"{job_dir}model_restart_{self.restart_epoch}.pth")
print(
f"-- Backing up as "
f"{job_dir}model_restart_{self.restart_epoch}_restarted.pth.",
flush=True,
)
shutil.copyfile(
f"{job_dir}model_restart_{self.restart_epoch}.pth",
f"{job_dir}model_restart_{self.restart_epoch}_restarted.pth",
)
else:
print("-- Initializing model from scratch.", flush=True)
self.model = models.initialize_model()
self.restart_epoch = 0
start_epoch = self.restart_epoch + 1
end_epoch = start_epoch + self.C.epochs
print("-- Defining optimizer.", flush=True)
self.optimizer = torch.optim.Adam(
params=self.model.parameters(),
lr=self.C.init_lr,
weight_decay=self.C.weight_decay,
)
return start_epoch, end_epoch