Troubleshooting
Benjamin Wilhelm edited this page Jan 25, 2019
·
4 revisions
Executing deep neural networks on big images is memory expensive. We have build techniques to make it possible on common modern hardware. See the wiki page of the specific Fiji command or KNIME workflows for detailed information.
If encounter problems which aren't described in this wiki, you can open an issue in one of the GitHub repositories:
| Name | OS | CPU | RAM | GPU | CUDA | Tested |
|---|---|---|---|---|---|---|
| Macbook Pro 2012 | macOS 10.12.6 | 2.7 GHz Core i7 | 16 GB | - | No | Fiji Plugins, KNIME workflows |
| XMG P406 | Xubuntu 17.04 | 2.6 GHz Core i7 | 32 GB | GeForce GTX 970M (3 GB) | Yes | Fiji Plugins |
| Paperspace ML-in-a-Box | Ubuntu 14.04 LTS | 8 x 2.6 GHz Xeon | 30 GB | Quadro P4000 | Yes | Fiji Plugins, KNIME workflows |
| Dell XPS 15 | Arch Linux | 2.8 GHz Core i7 | 32 GB | GeForce GTX 1050 (4 GB) | Yes | KNIME workflows |