search

google / in-silico-labeling

Code for In silico labeling: Predicting fluorescent labels in unlabeled images
Apache License 2.0
258 stars 80 forks source link

Trying to create a new label #6

Open Li-En-Good opened 1 year ago

Li-En-Good commented 1 year ago

Hi,

Hope you are doing well! I got this when trying to create a new label: KeyError: 'GFAP_CONFOCAL'

I see the channels you have are ('channel', ('DAPI_CONFOCAL', 'DAPI_WIDEFIELD', 'CELLMASK_CONFOCAL', 'TUJ1_WIDEFIELD', 'NFH_CONFOCAL', 'MAP2_CONFOCAL', 'ISLET_WIDEFIELD', 'DEAD_CONFOCAL', 'NEURITE_CONFOCAL')),

Is there a way to add new ones? Thanks!

emchristiansen commented 1 year ago

Hey Li,

This repo is reaaaaaaally out of date and is now basically just an artifact for replicating results from the paper. I suggest you find another image-to-image predictor to accomplish your task

Also, consider using JAX instead of TensorFlow - it's much nicer for almost everything, in my experience.

On Sat, Oct 1, 2022 at 5:15 PM Li-En-Good @.***> wrote:

Hi,

Hope you are doing well! I got this when trying to create a new label: KeyError: 'GFAP_CONFOCAL'

I see the channels you have are ('channel', ('DAPI_CONFOCAL', 'DAPI_WIDEFIELD', 'CELLMASK_CONFOCAL', 'TUJ1_WIDEFIELD', 'NFH_CONFOCAL', 'MAP2_CONFOCAL', 'ISLET_WIDEFIELD', 'DEAD_CONFOCAL', 'NEURITE_CONFOCAL')),

Is there a way to add new ones? Thanks!

— Reply to this email directly, view it on GitHub https://github.com/google/in-silico-labeling/issues/6, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABXDBD64DEC5MSOL7VFOHDWBBISJANCNFSM6AAAAAAQ2PUU3I . You are receiving this because you are subscribed to this thread.Message ID: @.***>

Li-En-Good commented 1 year ago

Thanks for the prompt reply and the suggestion. I really like your network design and wanted to try it. Do you happen to know other publications that uses JAX for a similar network architecture? I haven't really built a network before. I looked through the papers that cited yours and didn't find anything like it. Many thanks!