MouseLand / cellpose

a generalist algorithm for cellular segmentation with human-in-the-loop capabilities
https://www.cellpose.org/
BSD 3-Clause "New" or "Revised" License
1.24k stars 359 forks source link

running in a non internet environment #952

Open EricDeveaud opened 4 weeks ago

EricDeveaud commented 4 weeks ago

Hello,

I'm trying to install cellpose-3.0.8 to be run on a cluster whose compute nodes does not have internet access. Installation is done on a ReadOnly shared drive mounted on the nodes.

this mean that users won't be abble to download some required files. https://www.cellpose.org/static/images/cellpose_transparent.png https://www.cellpose.org/static/images/cellpose_gui.png

NB I can download thoose files and patch cellpose/gui/gui.py and cellpose/gui/gui3d.py to point to it.

models won't be abble to be downloaded by users.

after looking at docs/models.rst and models.py code I noticed some models described but not available eg docs/models.rst describe 3 cytoplasm models cyto3, cyto2and cyto. unfortunatly

https://www.cellpose.org/models/cyto2
https://www.cellpose.org/models/cyto

returns a 500 error

same happens for nuclei

while models.py describe more models than docs/models.rst. eg neurips_* ones.

can you please provide a comprehensive list of all defaults models that cellpose3 can use ? deblur_*, denoise_*, size_* and // or an archive whith the complete set.

I will download the models and store on a local storage pointing it using CELLPOSE_LOCAL_MODELS_PATH

regards

Eric

carsen-stringer commented 3 weeks ago

this zip file is out of date but we will update it soon and list the models available in the readthedocs: https://github.com/MouseLand/cellpose?tab=readme-ov-file#download-of-pretrained-models

EricDeveaud commented 3 weeks ago

so far what I was abble to determine is that cellpose-3.0.8 requires at list

bact_fluor_cp3
bact_phase_cp3
cyto2_cp3
cyto3
cytotorch_0
deblur_cyto3
deblur_nuclei
deepbacs_cp3
denoise_cyto3
denoise_nuclei
livecell_cp3
neurips_cellpose_default
neurips_cellpose_transformer
neurips_grayscale_cyto2
nucleitorch_0
size_cyto3.npy
size_nucleitorch_0.npy
tissuenet_cp3
transformer_cp3
upsample_cyto3
upsample_nuclei
yeast_BF_cp3
yeast_PhC_cp3

can you confirm this is correct and thta I'm not missing any further file//model

regards

Eric