I am running the is3g notebook on our HPC environment where it works well without the nuclei location data. As soon as I add the nuclei positions and try to run the segmentation I receive the following message:
The kernel for 02-spatial_transcriptomics/01-dataanalysis/08-ISS_processing/ISS_aging/4-ISS_spot_cleanup_is3g-subset.ipynb appears to have died. It will restart automatically.
The computing environment I am running this in is either:
- Large (28 core, 180GB RAM) or
- XLarge (42 core, 270GB RAM)
both with A100 Nvidia GPU support. As the name of the subset.ipynb suggests I already tried to subset the data to see whether the data size could be an issue, but the kernel still dies even though trying to run the segmentation on only one tile of the data.
Attached is the notebook of the subset of the data - not sure if the data is still too large?
4-ISS_spot_cleanup_is3g-subset.ipynb.zip
Please let me know if you need additional details.
I am running the is3g notebook on our HPC environment where it works well without the nuclei location data. As soon as I add the nuclei positions and try to run the segmentation I receive the following message:
The kernel for 02-spatial_transcriptomics/01-dataanalysis/08-ISS_processing/ISS_aging/4-ISS_spot_cleanup_is3g-subset.ipynb appears to have died. It will restart automatically.The computing environment I am running this in is either:
both with A100 Nvidia GPU support. As the name of the subset.ipynb suggests I already tried to subset the data to see whether the data size could be an issue, but the kernel still dies even though trying to run the segmentation on only one tile of the data.
Attached is the notebook of the subset of the data - not sure if the data is still too large?
4-ISS_spot_cleanup_is3g-subset.ipynb.zip
Please let me know if you need additional details.