SC-shendazt

Results 25 comments of SC-shendazt

Hi Dr.HuguesTHOMAS Unfortunately, it doesn't seem to have any performance improvements, so I added EDGECONV and self -- Attention

Dr.HuguesTHOMAS Thank you very much for your quick answer, I will follow your suggestions to improve this block

Hello mr.Thomas.Thank you for your wonderful work!!And I have met a question: self.all_splits=[0,1,2,3,4,5,6,7,8,9] if my validation is:3,5,6 how can i set the :sel.validation_split =[?]

> Thank you very much. I'll have a try first

@GeoSur “”in_radius or first_subsampling_dl “” Perhaps you could try different values for these two parameters eg:in_radius=16 first_subsampling_dl=0.2~0.X

![QQ图片20220328181827](https://user-images.githubusercontent.com/74888508/160378006-db1a61d1-8fef-4260-b519-2a89bc488747.png) ![QQ图片20220328181855](https://user-images.githubusercontent.com/74888508/160378027-775cdc59-9653-45f2-b2bb-b9195a451dca.png) Hello mr.Thomas.,And I have met a new question,Could you give me some advice?

Thanks for your reply.!

![QQ图片20220620230900](https://user-images.githubusercontent.com/74888508/174631839-610c06c7-36db-4fd6-aed6-4db0334397bb.png) Thank you very much for your quick reply. I find that this problem will not occur if I add this code. I wonder if I am correct in modifying...

> Both strategies would work indeed Thank you very much

> If I understand correctly, you want to concatenate the features from all layers as a final feature, instead of only using the N1 x 64 last feature, right? >...