convnext_atto

holocron.models.convnext_atto(pretrained: bool = False, checkpoint: Checkpoint | None = None, progress: bool = True, **kwargs: Any) → ConvNeXt[source]

ConvNeXt-Atto variant of Ross Wightman inspired by “A ConvNet for the 2020s”

Parameters:

  • pretrained – If True, returns a model pre-trained on ImageNette

  • checkpoint – If specified, the model’s parameters will be set to the checkpoint’s values

  • progress – If True, displays a progress bar of the download to stderr

  • kwargs – keyword args of _convnext

Returns:

classification model

Return type:

torch.nn.Module

class holocron.models.ConvNeXt_Atto_Checkpoint(value)[source]

The model builder above accepts the following values as the checkpoint parameter. ConvNeXt_Atto_Checkpoint.DEFAULT is equivalent to ConvNeXt_Atto_Checkpoint.IMAGENETTE

ConvNeXt_Atto_Checkpoint.IMAGENETTE:

Also available as ConvNeXt_Atto_Checkpoint.DEFAULT.

dataset

imagenette

top1-accuracy

0.8759

top5-accuracy

0.9832

url

link

sha256

f38217e7361060e6

size

12.9MB

num_params

3.4M

categories

tench, English springer, cassette player, … (7 omitted)

input_shape

(3, 224, 224)

mean

(0.485, 0.456, 0.406)

std

(0.229, 0.224, 0.225)

interpolation

InterpolationMode.BILINEAR

commit

d4a5999