HenriquesLab / ZeroCostDL4Mic

ZeroCostDL4Mic: A Google Colab based no-cost toolbox to explore Deep-Learning in Microscopy
MIT License
566 stars 130 forks source link

Stardist 2D prediction error output , 1000*1000 with 16bit #257

Open mdoaskndfweon opened 1 year ago

mdoaskndfweon commented 1 year ago
          hello, sorry for that confusing , I had already solve above those problem, but I have question relate my result for the nuclei segmentation , suppose that input image size is 1000,
Screenshot 2023-04-01 at 6 42 41 PM

this is the test result ,

Screenshot 2023-04-01 at 6 43 42 PM

for example , right hand side of the grapth is segmentation result, but when I try to predict new image without label,

Screenshot 2023-04-01 at 6 45 31 PM

the result are showed below

Screenshot 2023-04-01 at 6 45 07 PM

you see, totally differnt from the test data, but i want the format look like the same for the test data format.

my input block size x and y is 756, min overlap is256, When I decrease of block size or decrease of lin overlap, it show error and can't output the grapth

Originally posted by @mdoaskndfweon in https://github.com/HenriquesLab/ZeroCostDL4Mic/issues/256#issuecomment-1493164463

guijacquemet commented 1 year ago

Hi, Thanks for reaching out! I am not sure I understand your question. What do you mean by "!suppose that input image size is 1000" ?

Cheers Guillaume