Busy life

Results 6 comments of Busy life

> The second question. num_queries = 1 means 1query image every way As I have mentioned, there are 5 images (way) per class (shot) when num_queries = 1

> After taking a look at everyone's questions, I would like to ask, is it necessary to run the code? Not necessary. Just take a quick glance at the paper...

I figure out that you are using the data format from DeepEMD where all images are 224*224.

> 您好, > > 感谢您对于我们工作的兴趣。 > > 1. 这个项目包含的代码使用的是是ResNet-12。对应的结果在ECCV20论文的supplementary里面。 > 2. 这个项目使用的ResNet-12和MetaOptNet里的一样,channel的数量是64-160-320-640。MTL原始论文中ResNet-12的channel的数量是64-128-256-512。 > 3. 由于一直再忙别的项目,之前的代码一直没有找到时间更新。我们会尽快更新hard task,多谢提醒。 您好,感谢您的开源。 我尝试使用您提供的关于tieredImageNet的ResNet12预训练模型来fine-tune, 但模型似乎是没有经过预训练的(fine-tune只有20%的准确率),您能确认一下预训练模型吗? 谢谢!

> > 作者你好!请教一个问题,在测试的时候论文是将前75个有着最大energy score的query作为unknow,但如果是测试一张query,如何判断它是unknow呢?很多论文是采用阈值来判断unknow,而这里直接用前75个作为unknow,实际上引入了先验信息,这种测试方式我很疑惑到底合不合理。 > > I have the same question for the author. During the evaluation, author used the first 75 samples as unknown, that means author make use...

> @LoveMiki Hi, did u solve the problem? :) Sorry, I just gave up reproducing the results and preferred not to follow their work.