zzzzzz

https://github.com/yinboc/few-shot-meta-baseline/issues/8 你可以参考作者的这个回答。看起来是稍微调几步会好点(Fig 2中从epoch 1开始)，调多了就不好了。 所以meta-learning还是有用的，不过不能用太多步。

@YuxianMeng I seems that I also need use new train.objects.mmap.splitao.mmap? And there is no space left on my device. Can I still run normally after I delete origin_dir/train_images?

@YuxianMeng OK. Thank you a lot.

Your academic taste needs to improve.

Thanks for your reply. I ran it once under inter 5-way 1-shot setting and the results looked very bad. 2022-09-20 22:57:35 INFO: - span_f1 = 0.7218073781712385 2022-09-20 22:57:35 INFO: -...

I understand the performance will drop, but it performs poorly.

Thanks. The new result seems to be correct. 2022-09-24 20:43:14 INFO: - ***** Eval results inter-test ***** 2022-09-24 20:43:14 INFO: - f1 = 0.6104350036041772 2022-09-24 20:43:14 INFO: - f1_threshold =...

I am very sorry to interrupt again. Why is the performance in 5-shot worse than 1-shot after ablating fine-tuning in meta-test? For example, the F1 under inter 5way-5shot is about...

Thanks. Hope you have a good day.

对，是这样的，应该有点问题