cellpose
cellpose copied to clipboard
MouseLand
[BUG] can't use restore function with CLI
Describe the bug Can't run a restore from CLI
To Reproduce
- fresh install cellpose 3.1.0
mamba create -n cellpose-310 python=3.8 -y
mamba activate cellpose-310
mamba install pytorch pytorch-cuda=11.8 -c pytorch -c nvidia -y
pip install cellpose==3.1.0 scikit-image
- use the line adapted from from documentation https://cellpose.readthedocs.io/en/latest/restore.html#command-line-usage ,
python -Xutf8 -m cellpose --dir C:\Users\guiet\AppData\Local\Temp\cellposeTemp --model_type cyto3 --restore_type denoise_cyto3 --diameter 30 --verbose --save_tif --no_npy --use_gpu
or
python -Xutf8 -m cellpose --dir C:\Users\guiet\AppData\Local\Temp\cellposeTemp --model_type cyto3 --restore_type denoise_cyto3 --diameter 25 --chan2_restore --chan 1 --chan2 1
get error : __main__.py: error: unrecognized arguments: --model_type cyto3
NOTE : a minimal running line for cellpose segmentation without the restoration :
python -Xutf8 -m cellpose --dir C:\Users\guiet\AppData\Local\Temp\cellposeTemp --pretrained_model cyto3 --diameter 30 --verbose --save_tif --no_npy --use_gpu
If I remove --model_type cyto3 , like :
python -Xutf8 -m cellpose --dir C:\Users\guiet\AppData\Local\Temp\cellposeTemp --pretrained_model cyto3 --chan 0 --chan2 0 --diameter 0 --verbose --save_tif --no_npy --use_gpu --restore_type denoise_cyto3
I get the error :
2025-02-11 09:26:49,094 [INFO] WRITING LOG OUTPUT TO C:\Users\guiet\.cellpose\run.log
2025-02-11 09:26:49,095 [INFO]
cellpose version: 3.1.0
platform: win32
python version: 3.8.20
torch version: 2.4.1
2025-02-11 09:26:49,436 [INFO] ** TORCH CUDA version installed and working. **
2025-02-11 09:26:49,437 [INFO] >>>> using GPU (CUDA)
2025-02-11 09:26:49,446 [INFO] >>>> running cellpose on 1 images using chan_to_seg GRAY and chan (opt) NONE
2025-02-11 09:26:49,448 [INFO] >> denoise_cyto3 << model set to be used
D:\conda\conda-envs\cellpose-310\lib\site-packages\cellpose\resnet_torch.py:271: FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
state_dict = torch.load(filename, map_location=device)
2025-02-11 09:26:49,717 [INFO] >>>> model diam_mean = 30.000 (ROIs rescaled to this size during training)
2025-02-11 09:26:49,718 [INFO] >> cyto3 << model set to be used
2025-02-11 09:26:49,783 [INFO] >>>> loading model C:\Users\guiet\.cellpose\models\cyto3
2025-02-11 09:26:49,859 [INFO] >>>> model diam_mean = 30.000 (ROIs rescaled to this size during training)
2025-02-11 09:26:49,859 [INFO] >>>> cannot auto-estimate diameter for image restoration
Traceback (most recent call last):
File "D:\conda\conda-envs\cellpose-310\lib\runpy.py", line 194, in _run_module_as_main
return _run_code(code, main_globals, None,
File "D:\conda\conda-envs\cellpose-310\lib\runpy.py", line 87, in _run_code
exec(code, run_globals)
File "D:\conda\conda-envs\cellpose-310\lib\site-packages\cellpose\__main__.py", line 358, in <module>
main()
File "D:\conda\conda-envs\cellpose-310\lib\site-packages\cellpose\__main__.py", line 195, in main
diameter = model.diam_labels
AttributeError: 'CellposeDenoiseModel' object has no attribute 'diam_labels'
Thank you for your help,
Best,
Romain
one needs to specify the diameter value, it can't be 0! the line :
python -Xutf8 -m cellpose --dir C:\Users\guiet\AppData\Local\Temp\cellposeTemp --pretrained_model cyto3 --chan 0 --chan2 0 --diameter 30 --verbose --save_tif --no_npy --use_gpu --restore_type denoise_cyto3
works
Cheers
R
