I know this has been in various threads such as issues #292, but we are running Suite2P on a standard 1P fluorescent microscope to look at ex vivo cultures. Currently the masks we generate with Cellpose and with Suite2P don't align, similar to what was noted in #292.
That said, I think I can solve this by better training my custom Cellpose model. Can you let me know how Suite2P generates the average image for identifying cells? I realized that my Cellpose models are being trained on individual frames, but I should instead average (in the same way that Suite2P is).
Thanks!
Reproduce the code example:
Currently running in GUI (will move to a notebook once I have this solved)
Describe the issue:
Hello,
I know this has been in various threads such as issues #292, but we are running Suite2P on a standard 1P fluorescent microscope to look at ex vivo cultures. Currently the masks we generate with Cellpose and with Suite2P don't align, similar to what was noted in #292.
That said, I think I can solve this by better training my custom Cellpose model. Can you let me know how Suite2P generates the average image for identifying cells? I realized that my Cellpose models are being trained on individual frames, but I should instead average (in the same way that Suite2P is).
Thanks!
Reproduce the code example:
Error message:
No response
Version information:
suite2p v0.14.4
Context for the issue:
No response