Inception-v3报错：RuntimeError: Calculated padded input size per channel: (3 x 3). Kernel size: (5 x 5)

在PyTorch1.7.1环境下运行PyTorch官方提供的ImageNet预训练模型Inception-v3报错如下：RuntimeError: Calculated padded input size per channel: (3 x 3). Kernel size: (5 x 5). Kernel size can't be greater than actual input size报错

YulongYang1896

22794人浏览 · 2021-09-28 19:56:29

YulongYang1896 · 2021-09-28 19:56:29 发布

在PyTorch1.7.1环境下运行PyTorch官方提供的ImageNet预训练模型Inception-v3报错如下：

RuntimeError: Calculated padded input size per channel: (3 x 3). Kernel size: (5 x 5). Kernel size can't be greater than actual input size

报错原因是我混用了其他模型的预处理transform方法，导致图片大小与网络结构不匹配。解决办法是采用PyTorch官方提供的transform方法如下：https://pytorch.org/hub/pytorch_vision_inception_v3/

'''for inception-v3'''
    transform = transforms.Compose([
        transforms.Resize(299),
        transforms.CenterCrop(299),
        transforms.ToTensor(),
        transforms.Normalize(mean=[0.485, 0.456, 0.406],
                             std=[0.229, 0.224, 0.225])
    ])