RuntimeWarning: invalid value encountered in power

[ransona]

Hi

I get the following runtime error/warnings:

/home/adamranson/code/SRDTrans/test.py:215: RuntimeWarning: invalid value encountered in power
= aaaa * (np.sum(bbbb) / np.sum(aaaa)) ** 0.5
/home/adamranson/code/SRDTrans/test.py:227: RuntimeWarning: invalid value encountered in cast
output_img=np.clip(output_img, 0, 65535).astype('int32')

Any idea what the problem might be? It happens with specific tif files but not others but I didn't see any obvious difference in the raw data. The raw data is 16 bit signed.

Thanks
Adam

....[Model 1/1, E_20_Iter_3430.pth] [Stack 1/1, combined_plane7_ch0.tif] [Patch 5118/5121] [Time Cost: 531 s] [ETA: 0 s]
[Model 1/1, E_20_Iter_3430.pth] [Stack 1/1, combined_plane7_ch0.tif] [Patch 5119/5121] [Time Cost: 531 s] [ETA: 0 s]
[Model 1/1, E_20_Iter_3430.pth] [Stack 1/1, combined_plane7_ch0.tif] [Patch 5120/5121] [Time Cost: 531 s] [ETA: 0 s]
[Model 1/1, E_20_Iter_3430.pth] [Stack 1/1, combined_plane7_ch0.tif] [Patch 5121/5121] [Time Cost: 531 s] [ETA: 0 s]
test result saved in: /home/adamranson/data/Repository/ESPM115/2024-11-11_01_ESPM115/denoised_tifs/suite2p/plane7/denoised_tif/DataFolderIs_reg_tif_202412071200_ModelFolderIs_mixed_axon_soma_g8_202412022250/E_20_Iter_3430/combined_plane7_ch0_E_20_Iter_3430_output.tif
/home/adamranson/miniconda3/envs/srdtrans/lib/python3.8/site-packages/timm/models/layers/init.py:48: FutureWarning: Importing from timm.models.layers is deprecated, please import via timm.layers
warnings.warn(f"Importing from {name} is deprecated, please import via timm.layers", FutureWarning)
/home/adamranson/miniconda3/envs/srdtrans/lib/python3.8/site-packages/torch/functional.py:513: UserWarning: torch.meshgrid: in an upcoming release, it will be required to pass the indexing argument. (Triggered internally at ../aten/src/ATen/native/TensorShape.cpp:3609.)
return _VF.meshgrid(tensors, **kwargs) # type: ignore[attr-defined]
/home/adamranson/code/SRDTrans/test.py:140: 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.
denoise_generator.module.load_state_dict(torch.load(model_name)) # parallel
/home/adamranson/code/SRDTrans/test.py:215: RuntimeWarning: invalid value encountered in power
= aaaa * (np.sum(bbbb) / np.sum(aaaa)) ** 0.5
/home/adamranson/code/SRDTrans/test.py:227: RuntimeWarning: invalid value encountered in cast
output_img=np.clip(output_img, 0, 65535).astype('int32')

[ransona]

Hi
The problem is arising due to aaaa * (np.sum(bbbb) / np.sum(aaaa)) containing negative value (as does the raw tif) and aaaa * (np.sum(bbbb) / np.sum(aaaa)) ** 0.5 therefore being a sqrt of a negative number which gives a nan in numpy. If I subtract the min of the stack pixel values then it works - i.e. I get a final denoised output with no zero patches.
I assume the solution is to ensure tifs only contain positive values?
Thanks
Adam