mnasnet0_75

torchvision.models.mnasnet0_75(*, weights: Optional[MNASNet0_75_Weights] = None, progress: bool = True, **kwargs: Any) → MNASNet

MNASNet with depth multiplier of 0.75 from MnasNet: Platform-Aware Neural Architecture Search for Mobile paper.

Parameters:

• weights (MNASNet0_75_Weights, optional) – The pretrained weights to use. See MNASNet0_75_Weights below for more details, and possible values. By default, no pre-trained weights are used.

• progress (bool, optional) – If True, displays a progress bar of the download to stderr. Default is True.

• **kwargs – parameters passed to the torchvision.models.mnasnet.MNASNet base class. Please refer to the source code for more details about this class.

class torchvision.models.MNASNet0_75_Weights(value)

The model builder above accepts the following values as the weights parameter. MNASNet0_75_Weights.DEFAULT is equivalent to MNASNet0_75_Weights.IMAGENET1K_V1. You can also use strings, e.g. weights='DEFAULT' or weights='IMAGENET1K_V1'.

MNASNet0_75_Weights.IMAGENET1K_V1:

These weights were trained from scratch by using TorchVision’s new training recipe. Also available as MNASNet0_75_Weights.DEFAULT.

acc@1 (on ImageNet-1K)	71.18
acc@5 (on ImageNet-1K)	90.496
min_size	height=1, width=1
categories	tench, goldfish, great white shark, … (997 omitted)
recipe	link
num_params	3170208
GFLOPS	0.21
File size	12.3 MB

The inference transforms are available at MNASNet0_75_Weights.IMAGENET1K_V1.transforms and perform the following preprocessing operations: Accepts PIL.Image, batched (B, C, H, W) and single (C, H, W) image torch.Tensor objects. The images are resized to resize_size=[232] using interpolation=InterpolationMode.BILINEAR, followed by a central crop of crop_size=[224]. Finally the values are first rescaled to [0.0, 1.0] and then normalized using mean=[0.485, 0.456, 0.406] and std=[0.229, 0.224, 0.225].