search

BodenmillerGroup / ImcSegmentationPipeline

A pixel classification based multiplexed image segmentation pipeline
https://bodenmillergroup.github.io/ImcSegmentationPipeline/
MIT License
83 stars 35 forks source link

Images of type float must be between -1 and 1 #57

Closed StarXian closed 2 years ago

StarXian commented 3 years ago

When I use the V2, Cellprofiler: 1_prepare_ilastik, an error message pops up “Images of type float must be between -1 and 1” and this happens with only some of my images

votti commented 3 years ago

Hi StarXian, Thanks for reporting this - you run into an issue that I did not consider well enough.

I made now a branch that should have this fixed: https://github.com/BodenmillerGroup/ImcSegmentationPipeline/tree/feature-fix-rescaling

Can you try the 1_prepare_ilastik pipeline from this branch: https://github.com/BodenmillerGroup/ImcSegmentationPipeline/blob/feature-fix-rescaling/cp4_pipelines/3_measure_mask_basic.cppipe

Sorry for the issues and thanks for reporting if this works! Best

StarXian commented 3 years ago

Hi StarXian, Thanks for reporting this - you run into an issue that I did not consider well enough.

I made now a branch that should have this fixed: https://github.com/BodenmillerGroup/ImcSegmentationPipeline/tree/feature-fix-rescaling

Can you try the 1_prepare_ilastik pipeline from this branch: https://github.com/BodenmillerGroup/ImcSegmentationPipeline/blob/feature-fix-rescaling/cp4_pipelines/3_measure_mask_basic.cppipe

Sorry for the issues and thanks for reporting if this works! Best

Hi Vito, Thank you for your reply. It works very well. Perfect! Thank you so much.

By the way, I'd like to use the R implementation for the HistoCAT neightbourhood analysis, but the number from the CellProfiler output can't match with the number from histoCAT exported as the csv file, is there something wrong with my operation? The number from histoCAT is less than the CellProfiler output .

nilseling commented 2 years ago

This appears to be fixed. Please reopen if needed.