Skip to content

Latest commit

 

History

History
86 lines (71 loc) · 3.01 KB

pytorch_vision_squeezenet.md

File metadata and controls

86 lines (71 loc) · 3.01 KB
Raw
layout background-class body-class title summary category image author tags github-link github-id featured_image_1 featured_image_2 accelerator order
hub_detail
hub-background
hub
SqueezeNet
Alexnet-level accuracy with 50x fewer parameters.
researchers
squeezenet.png
Pytorch Team
vision
scriptable
pytorch/vision
squeezenet.png
no-image
cuda-optional
10
 
import torch
model = torch.hub.load('pytorch/vision:v0.6.0', 'squeezenet1_0', pretrained=True)
# or
# model = torch.hub.load('pytorch/vision:v0.6.0', 'squeezenet1_1', pretrained=True)
model.eval()

All pre-trained models expect input images normalized in the same way, i.e. mini-batches of 3-channel RGB images of shape (3 x H x W), where H and W are expected to be at least 224. The images have to be loaded in to a range of [0, 1] and then normalized using mean = [0.485, 0.456, 0.406] and std = [0.229, 0.224, 0.225].

Here's a sample execution.

# Download an example image from the pytorch website
import urllib
url, filename = ("https://github.com/pytorch/hub/raw/master/images/dog.jpg", "dog.jpg")
try: urllib.URLopener().retrieve(url, filename)
except: urllib.request.urlretrieve(url, filename)
# sample execution (requires torchvision)
from PIL import Image
from torchvision import transforms
input_image = Image.open(filename)
preprocess = transforms.Compose([
    transforms.Resize(256),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225]),
])
input_tensor = preprocess(input_image)
input_batch = input_tensor.unsqueeze(0) # create a mini-batch as expected by the model

# move the input and model to GPU for speed if available
if torch.cuda.is_available():
    input_batch = input_batch.to('cuda')
    model.to('cuda')

with torch.no_grad():
    output = model(input_batch)
# Tensor of shape 1000, with confidence scores over Imagenet's 1000 classes
print(output[0])
# The output has unnormalized scores. To get probabilities, you can run a softmax on it.
print(torch.nn.functional.softmax(output[0], dim=0))

Model Description

Model squeezenet1_0 is from the SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size paper

Model squeezenet1_1 is from the official squeezenet repo. It has 2.4x less computation and slightly fewer parameters than squeezenet1_0, without sacrificing accuracy.

Their 1-crop error rates on imagenet dataset with pretrained models are listed below.

Model structure Top-1 error Top-5 error
squeezenet1_0 41.90 19.58
squeezenet1_1 41.81 19.38

References