search

danfenghong / IEEE_TGRS_SpectralFormer

Danfeng Hong, Zhu Han, Jing Yao, Lianru Gao, Bing Zhang, Antonio Plaza, Jocelyn Chanussot. Spectralformer: Rethinking hyperspectral image classification with transformers, IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2021
219 stars 39 forks source link

picture color promble #11

Open Jasper0420 opened 2 years ago

Jasper0420 commented 2 years ago

老师我想请问一下，如何让图片结果呈现红框这种效果呢？我test的结果是蓝框效果 image

danfenghong commented 2 years ago

红框是原始数据的有标注的样本，黑色的背景是没有给出真值的地方，你的蓝筐效果是对每个点都做了预测，所以是全图的。

Jasper0420 @.***> 于2022年5月3日周二 23:23写道：

老师我想请问一下，如何让图片结果呈现红框这种效果呢？我test的结果是蓝框效果 [image: image] https://user-images.githubusercontent.com/63343279/166483486-074cebf1-f135-4b8c-9f8d-ba7f39219118.png

— Reply to this email directly, view it on GitHub https://github.com/danfenghong/IEEE_TGRS_SpectralFormer/issues/11, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFL2GZTYUPT36CHH7RPDGL3VIFAFNANCNFSM5U7HV42A . You are receiving this because you are subscribed to this thread.Message ID: @.***>

Jasper0420 commented 2 years ago

谢谢老师解答！请老师问如何只对有标注的样本进行预测呢？在代码中需要有哪些改动呢？谢谢

danfenghong commented 2 years ago

我们给出了训练和测试的样本集合，TR和TE。我们显示的量化精度就是只使用了测试样本进行计算的。

Jasper0420 @.***> 于2022年5月3日周二 23:58写道：

谢谢老师解答！请老师问如何只对有标注的样本进行预测呢？在代码中需要有哪些改动呢？谢谢

— Reply to this email directly, view it on GitHub https://github.com/danfenghong/IEEE_TGRS_SpectralFormer/issues/11#issuecomment-1116264346, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFL2GZSUWTUOVTRSKW7TNKTVIFELZANCNFSM5U7HV42A . You are receiving this because you commented.Message ID: @.***>

Jasper0420 commented 2 years ago

谢谢！