LiShenglana

Results 4 issues of LiShenglana

Does imgaug support the exact same augmentation for images within a batch, e.g. iaa.Sequential([iaa.Fliplr(0.4),....]) What is the best way to do this, with images within a batch flipped or not...

您好!请问您在训练过程中有加载ResNet50的预训练参数吗?因为代码中有一个位置好像修改了卷积核大小,没办法加载预训练好的模型。我用的训练数据集较小,想请教下这个问题怎么解决? 感谢!

This paper studies a good work on self-supervised tracking. How are the labels in json generated during the model training process?

您好!请问代码中提到的模型的两个分类标签如何获得?代码中的self._read_meta函数没有找到定义的地方,所以有点困惑,期待收到您的回复,谢谢! ![CZ}9}9_VE5{ZYO{CSN4G{WV](https://github.com/zhang-pengyu/HMFT/assets/106595830/447156ee-484c-4b07-821b-1dd7497f9be3) ![`K495Z`(Y00P%6Z481V`X L](https://github.com/zhang-pengyu/HMFT/assets/106595830/ba836470-d601-49c6-8044-9b98f0f7d65c)