超参数Tuning
optuna
示例：
import optuna
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
def objective(trial):
lr = trial.suggest_float("lr", 1e-4, 1e-2, log=True)
m = MyCNN().to(device)
optimizer = optim.Adam(m.parameters(), lr=lr)
criterion = nn.CrossEntropyLoss()
for epoch in range(5):
m.train()
for inputs, labels in trainloader:
inputs, labels = inputs.to(device), labels.to(device)
optimizer.zero_grad()
outputs = m(inputs)
loss = criterion(outputs, labels)
loss.backward()
optimizer.step()
# EVAL
m.eval()
correct = 0
total = 0
with torch.no_grad():
for inputs, labels in testloader:
inputs, labels = inputs.to(device), labels.to(device)
outputs = m(inputs)
_, predicted = torch.max(outputs, 1)
total += labels.size(0)
correct += (predicted == labels).sum().item()
return correct / total
def print_progress(study, trial):
print(f"Trial {trial.number}: lr={trial.params['lr']:.6f}, acc={trial.value:.4f}")
study = optuna.create_study(direction="maximize")
study.optimize(objective, n_trials=10, callbacks=[print_progress])
print("Best lr:", study.best_params)
。