[BUG] ask for one-click_nuclei but modelt set to denoise_cyto3

[pamaiuri]

Describe the bug
Both on the CellPose gui or from command line, when I ask for image restoration 'one-click' nuclei ( --restore_type 'oneclick_nuclei' on the command line) I got no error but it seams to load the default metod.

This is the log during the operation on the gui:
one-click_nuclei
2025-01-22 16:41:09,502 [INFO] >> denoise_cyto3 << model set to be used
2025-01-22 16:41:09,503 [INFO] >>>> using CPU
2025-01-22 16:41:09,503 [INFO] >>>> using CPU
2025-01-22 16:41:09,706 [INFO] >>>> model diam_mean = 30.000 (ROIs rescaled to this size during training)
GUI_INFO: channels: [0, 0]
GUI_INFO: normalize_params: {'lowhigh': None, 'percentile': [1.0, 99.0], 'normalize': True, 'norm3D': True, 'sharpen_radius': 0, 'smooth_radius': 0, 'tile_norm_blocksize': 0, 'tile_norm_smooth3D': 1, 'invert': False}
GUI_INFO: diameter (before upsampling): 30.0
(1, 2048, 2048, 1)
(2048, 2048, 1)
2025-01-22 16:41:19,777 [INFO] one-click_nuclei finished in 10.278 sec

This the one from command line (--restore_type 'oneclick_nuclei'):
2025-01-22 15:53:07,815 [INFO] >>>> running cellpose on 1 images using chan_to_seg GRAY and chan (opt) NONE
2025-01-22 15:53:07,815 [INFO] >> denoise_cyto3 << model set to be used
/home/PM_Application/CellPose/cellpose/lib/python3.12/site-packages/cellpose/resnet_torch.py:275: 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=torch.device("cpu"))
2025-01-22 15:53:08,019 [INFO] >>>> model diam_mean = 30.000 (ROIs rescaled to this size during training)
2025-01-22 15:53:08,019 [INFO] >> nuclei << model set to be used
2025-01-22 15:53:08,060 [INFO] >>>> loading model /home/paolo/.cellpose/models/nucleitorch_0
2025-01-22 15:53:08,128 [INFO] >>>> model diam_mean = 17.000 (ROIs rescaled to this size during training)
2025-01-22 15:53:08,128 [INFO] >>>> using diameter 20.000 for all images

Is it just wrong message or for some reason I did a mistake and I didn't manage to set the desired model?

Thank you for your help.

[pamaiuri]

update, from command line (--restore_type 'oneclick_nuclei') we got now this:

2025-01-24 15:33:05,812 [INFO] Downloading: "https://www.cellpose.org/models/oneclick_nuclei" to /home/maky/.cellpose/models/oneclick_nuclei
100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 25.4M/25.4M [00:03<00:00, 8.32MB/s]
2025-01-24 15:33:09,575 [INFO] >>>> running cellpose on 1 images using chan_to_seg GRAY and chan (opt) NONE
2025-01-24 15:33:09,576 [INFO] >> denoise_cyto3 << model set to be used

thank you all for help.