inception_v3

torchvision.models.quantization.inception_v3(*, weights: Optional[Union[Inception_V3_QuantizedWeights, Inception_V3_Weights]] = None, progress: bool = True, quantize: bool = False, **kwargs: Any) → QuantizableInception3

Inception v3 model architecture from Rethinking the Inception Architecture for Computer Vision.

Note

Important: In contrast to the other models the inception_v3 expects tensors with a size of N x 3 x 299 x 299, so ensure your images are sized accordingly.

Note

Note that quantize = True returns a quantized model with 8 bit weights. Quantized models only support inference and run on CPUs. GPU inference is not yet supported.

Parameters:

• weights (Inception_V3_QuantizedWeights or Inception_V3_Weights, optional) – The pretrained weights for the model. See Inception_V3_QuantizedWeights 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.

• quantize (bool, optional) – If True, return a quantized version of the model. Default is False.

• **kwargs – parameters passed to the torchvision.models.quantization.QuantizableInception3 base class. Please refer to the source code for more details about this class.

class torchvision.models.quantization.Inception_V3_QuantizedWeights(value)

The model builder above accepts the following values as the weights parameter. Inception_V3_QuantizedWeights.DEFAULT is equivalent to Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_V1. You can also use strings, e.g. weights='DEFAULT' or weights='IMAGENET1K_FBGEMM_V1'.

Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_V1:

These weights were produced by doing Post Training Quantization (eager mode) on top of the unquantized weights listed below. Also available as Inception_V3_QuantizedWeights.DEFAULT.

acc@1 (on ImageNet-1K)	77.176
acc@5 (on ImageNet-1K)	93.354
num_params	27161264
min_size	height=75, width=75
categories	tench, goldfish, great white shark, … (997 omitted)
backend	fbgemm
recipe	link
unquantized	Inception_V3_Weights.IMAGENET1K_V1
GIPS	5.71
File size	23.1 MB

The inference transforms are available at Inception_V3_QuantizedWeights.IMAGENET1K_FBGEMM_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=[342] using interpolation=InterpolationMode.BILINEAR, followed by a central crop of crop_size=[299]. 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].

class torchvision.models.Inception_V3_Weights(value)

The model builder above accepts the following values as the weights parameter. Inception_V3_Weights.DEFAULT is equivalent to Inception_V3_Weights.IMAGENET1K_V1. You can also use strings, e.g. weights='DEFAULT' or weights='IMAGENET1K_V1'.

Inception_V3_Weights.IMAGENET1K_V1:

These weights are ported from the original paper. Also available as Inception_V3_Weights.DEFAULT.

acc@1 (on ImageNet-1K)	77.294
acc@5 (on ImageNet-1K)	93.45
num_params	27161264
min_size	height=75, width=75
categories	tench, goldfish, great white shark, … (997 omitted)
recipe	link
GFLOPS	5.71
File size	103.9 MB

The inference transforms are available at Inception_V3_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=[342] using interpolation=InterpolationMode.BILINEAR, followed by a central crop of crop_size=[299]. 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].