composer.functional

Algorithms can be used directly through our functions-based API.

from composer import functional as CF
from torchvision import models

model = models.resnet50()

# replace some layers with blurpool
CF.apply_blurpool(model)
# replace some layers with squeeze-excite
CF.apply_se(model, latent_channels=64, min_channels=128)

functional.augment_and_mix	Perform augmentations.
functional.apply_blurpool	Applies BlurPool algorithm to the provided model.
functional.colout	Drops random rows and columns from a single image.
functional.cutout	Implements CutOut augmentation from https://arxiv.org/abs/1708.04552 on the batch level.
functional.smooth_labels	Shrinks targets towards a prior distribution to counteract label noise.
functional.freeze_layers	Implements the layer freezing algorithm.
functional.mixup_batch	Implements mixup on a single batch of data.
functional.resize_inputs	Resize inputs and optionally outputs by cropping or interpolating.
functional.randaugment	Perform augmentations.
functional.scale_scheduler
functional.selective_backprop	Select a subset of the batch on which to learn.
functional.apply_se	Adds Squeeze-and-Excitation <https://arxiv.org/abs/1709.01507>`_ (SE) blocks after the Conv2d layers of a neural network.