HantingChen

> 您好，我可以理解公式(9) 计算addnet输出方差的公式 (1-2/π)...，但是在其他版本中为什么会变成sqrt(π/2)....的形式呢？希望得到您的指点，谢谢 sqrt(π/2)....的形式为计算有误，请参考正确的版本

Sorry for the misleading. In fact, we measure the "depth" for a network using the number of its non-linear layers in its main branch instead of the number of conv...

你好，可能有以下几个原因：1.检查下载的权重是否损坏，尝试重新下载一份权重。2.使用的pytroch版本和requirements里需求的版本没有对应上。

Thanks for the nice concern. Since the BatchNorm will normalize the input features, it is almost the same for setting bias=True and bias=False. Therefore, both settings are fine.

你好，https://github.com/huawei-noah/VanillaNet/tree/main/object_detection 有在coco上实例分割的代码，可以参考。

hanks for your interest in our work! The "/path/to/imagenet-1k/" in the script actually represents the directory where you've downloaded the ImageNet dataset. Please refer to the "Dataset Preparation" section in...

> Hello, I've been testing the speed of VanillaNet recently. I tried VanillaNet6 with an image size of 128_128, which is smaller than 224_224. Using tensorrt for inference, the speed...

> > > Hello, I've been testing the speed of VanillaNet recently. I tried VanillaNet6 with an image size of 128_128, which is smaller than 224_224. Using tensorrt for inference,...

> @HantingChen The link seems to have expired ! Does the pretrained model translate well into real world dataset? Please refer to https://gitee.com/mindspore/models/tree/master/research/cv/IPT The pretrained model traslate well into real...

Thank you for your question! Our aim is to maximize the parallelism of the network and reduce the number of layers as much as possible to achieve better speed on...