Ruijia Wu

Looking Forward to SwinV2!

16

Thanks for the great work! I find the Swin V2 on arxiv just now, looking forward to the code of v2!

Thank you for what you have done! I'm confused about the translate matrix .here are the specific question: (1)Is the original focal length=1(i.e. if elevation=0deg) ? (2)Why the translate matrix...

如何阶段性保存模型？在训练过程中valset的coco_eval的AP一直是0？total_loss较大

13

作者，你好，我有几个问题想请教一下： 1. 我发现目前工程只在训练完全结束后才会保存模型，请问如何阶段性保存模型呢？我通过pip install安装了detectron2，随后在detectron2.engine.defaults.py中的DefaultTrainer增加train函数（以期覆盖TrainerBase中的train函数），具体代码如下（基于TrainerBase.train(), 增加了一行print, 以及阶段性保存模型的代码）： ` def train(self, start_iter: int, max_iter: int): """ Args: start_iter, max_iter (int): See docs above """ logger = logging.getLogger(__name__) logger.info("Starting training from iteration {}".format(start_iter))...

Does anybody encounter the same issue as mine? my custom dataset: train:755, test: 83 I've tried 2 plans: 1. vgg11 as backbone: the test result of epoch 10 is: the...

Thanks for your great work. Could you please tell me the specific version used in huggingface demo? That is, the base model? , and the pulid model ?. Looking forward...