Mixed precision training¶
Catalyst support a variety of backends for mixed precision training.
After PyTorch 1.6 release, it’s possible to use AMP natively inside torch.amp
package.
Suppose you have the following pipeline with Linear Regression:
import torch
from torch.utils.data import DataLoader, TensorDataset
from catalyst.dl import SupervisedRunner
# data
num_samples, num_features = int(1e4), int(1e1)
X, y = torch.rand(num_samples, num_features), torch.rand(num_samples)
dataset = TensorDataset(X, y)
loader = DataLoader(dataset, batch_size=32, num_workers=1)
loaders = {"train": loader, "valid": loader}
# model, criterion, optimizer, scheduler
model = torch.nn.Linear(num_features, 1)
criterion = torch.nn.MSELoss()
optimizer = torch.optim.Adam(model.parameters())
scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, [3, 6])
# model training
runner = SupervisedRunner()
runner.train(
model=model,
criterion=criterion,
optimizer=optimizer,
scheduler=scheduler,
loaders=loaders,
logdir="./logdir",
valid_loader="valid",
valid_metric="loss",
minimize_valid_metric=True,
num_epochs=8,
verbose=True,
)
Torch AMP¶
If you would like to use native AMP support, you could do the following:
import torch
from torch.utils.data import DataLoader, TensorDataset
from catalyst.dl import SupervisedRunner
# data
num_samples, num_features = int(1e4), int(1e1)
X, y = torch.rand(num_samples, num_features), torch.rand(num_samples)
dataset = TensorDataset(X, y)
loader = DataLoader(dataset, batch_size=32, num_workers=1)
loaders = {"train": loader, "valid": loader}
# model, criterion, optimizer, scheduler
model = torch.nn.Linear(num_features, 1)
criterion = torch.nn.MSELoss()
optimizer = torch.optim.Adam(model.parameters())
scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, [3, 6])
# model training
runner = SupervisedRunner()
runner.train(
model=model,
criterion=criterion,
optimizer=optimizer,
scheduler=scheduler,
loaders=loaders,
logdir="./logdir",
valid_loader="valid",
valid_metric="loss",
minimize_valid_metric=True,
num_epochs=8,
verbose=True,
fp16=True,
)
You could also check out the example above in this Google Colab notebook
If you haven’t found the answer for your question, feel free to join our slack for the discussion.