weixin_39804523
weixin_39804523
2020-12-01 19:07

Bug found by user and corrected. Specifically, the gaussian random fi…

Bug found by user and corrected. Specifically, the gaussian random fields had asymmetric variance when the X and Y dimensions were not symmetrical. This led to rippling type pattern in the gaussian fields. Fixed by upsampling the volume into a symmetrical dimensionality and then cropping after generation.

该提问来源于开源项目:brainiak/brainiak

  • 点赞
  • 写回答
  • 关注问题
  • 收藏
  • 复制链接分享
  • 邀请回答

6条回答

  • weixin_39628268 weixin_39628268 5月前

    It seems that most checks have passed except for those annoying ones without report. Do you want me to review this PR? :)

    点赞 评论 复制链接分享
  • weixin_39804523 weixin_39804523 5月前

    Yes please, do look. Shouldn't be too much. Not sure why Travis is failing but also it won't let me log in to retest the job. Will review after you have been through the PR

    点赞 评论 复制链接分享
  • weixin_39628268 weixin_39628268 5月前

    Hi , can you explain a bit more of the bug? Is it that if the first 3 dimensions are not equal, then the fft of white noise has different number of grids on the 3 dimensions in frequency domain, then the lowest spatial frequencies represented in each dimension are different?

    点赞 评论 复制链接分享
  • weixin_39804523 weixin_39804523 5月前

    Yes, that is my understanding of the bug. Then I think that ultimately means the variance is 'truncated' into different ranges

    点赞 评论 复制链接分享
  • weixin_39804523 weixin_39804523 5月前

    , thank you so much! , should you do it or I?

    点赞 评论 复制链接分享
  • weixin_39857513 weixin_39857513 5月前

    , you can certainly merge it yourself. Try to set a one-line title please, such as: fmrisim: Fix asymmetric variance bug

    点赞 评论 复制链接分享

相关推荐