# Load a pre-trained model model = torchvision.models.resnet50(pretrained=True)
# Disable gradient computation since we're only doing inference with torch.no_grad(): features = model(input_data)
# Example input input_data = torch.randn(1, 3, 224, 224) # 1 image, 3 channels, 224x224 pixels
# Load a pre-trained model model = torchvision.models.resnet50(pretrained=True)
# Disable gradient computation since we're only doing inference with torch.no_grad(): features = model(input_data)
# Example input input_data = torch.randn(1, 3, 224, 224) # 1 image, 3 channels, 224x224 pixels
