not working on large 2D images

Hi
I tried running a model on a large 2D image (4856 x 3672), and I got this error:

Model loading: 0
Engines loading: 0
Set size: 0
Title panel: 21
Search bar: 11
Content panel: 18
Footer: 1
done
[SERVICE-0] {"task":"38b52de7-6e96-4a94-9352-f07c3d66784a","requestType":"EXECUTE","inputs":{"modelFolder":".Fiji.app\\models\\2D_enteric_neuron.bioimage.io.model"},"script":"loadModel"}
[SERVICE-0] {"responseType":"LAUNCH","task":"38b52de7-6e96-4a94-9352-f07c3d66784a"}
[WORKER-0] 2025-02-24 02:07:13.438356: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_100.dll'; dlerror: cudart64_100.dll not found
[WORKER-0] 2025-02-24 02:07:13.438384: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
[WORKER-0] 2025-02-24 02:07:13.481728: I tensorflow/cc/saved_model/reader.cc:31] Reading SavedModel from: C:\Users\Pradeep\Desktop\Fiji.app\models\2D_enteric_neuron.bioimage.io.model
[WORKER-0] 2025-02-24 02:07:13.503051: I tensorflow/cc/saved_model/reader.cc:54] Reading meta graph with tags { serve }
[WORKER-0] 2025-02-24 02:07:13.541302: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
[WORKER-0] 2025-02-24 02:07:13.545032: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library nvcuda.dll
[WORKER-0] 2025-02-24 02:07:13.563714: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1618] Found device 0 with properties: 
[WORKER-0] name: NVIDIA GeForce RTX 3080 major: 8 minor: 6 memoryClockRate(GHz): 1.755
[WORKER-0] pciBusID: 0000:2d:00.0
[WORKER-0] 2025-02-24 02:07:13.564463: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudart64_100.dll'; dlerror: cudart64_100.dll not found
[WORKER-0] 2025-02-24 02:07:13.565004: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cublas64_100.dll'; dlerror: cublas64_100.dll not found
[WORKER-0] 2025-02-24 02:07:13.565512: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cufft64_100.dll'; dlerror: cufft64_100.dll not found
[WORKER-0] 2025-02-24 02:07:13.566010: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'curand64_100.dll'; dlerror: curand64_100.dll not found
[WORKER-0] 2025-02-24 02:07:13.566501: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cusolver64_100.dll'; dlerror: cusolver64_100.dll not found
[WORKER-0] 2025-02-24 02:07:13.566979: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cusparse64_100.dll'; dlerror: cusparse64_100.dll not found
[WORKER-0] 2025-02-24 02:07:13.567477: W tensorflow/stream_executor/platform/default/dso_loader.cc:55] Could not load dynamic library 'cudnn64_7.dll'; dlerror: cudnn64_7.dll not found
[WORKER-0] 2025-02-24 02:07:13.567487: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1641] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
[WORKER-0] Skipping registering GPU devices...
[WORKER-0] 2025-02-24 02:07:13.679874: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1159] Device interconnect StreamExecutor with strength 1 edge matrix:
[WORKER-0] 2025-02-24 02:07:13.679892: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1165]      0 
[WORKER-0] 2025-02-24 02:07:13.679896: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1178] 0:   N 
[WORKER-0] 2025-02-24 02:07:13.798778: I tensorflow/cc/saved_model/loader.cc:202] Restoring SavedModel bundle.
[WORKER-0] 2025-02-24 02:07:14.124703: I tensorflow/cc/saved_model/loader.cc:311] SavedModel load for tags { serve }; Status: success. Took 642971 microseconds.
[SERVICE-0] {"responseType":"COMPLETION","task":"38b52de7-6e96-4a94-9352-f07c3d66784a"}
java.lang.NullPointerException
	at io.bioimage.modelrunner.bioimageio.tiling.TileCalculator.calculateByteSizeOfAffectedOutput(TileCalculator.java:254)
	at io.bioimage.modelrunner.bioimageio.tiling.TileCalculator.checkOutputSize(TileCalculator.java:178)
	at io.bioimage.modelrunner.bioimageio.tiling.TileCalculator.getOptimalTileSize(TileCalculator.java:124)
	at io.bioimage.modelrunner.model.BioimageIoModel.run(BioimageIoModel.java:329)
	at deepimagej.Runner.run(Runner.java:138)
	at deepimagej.gui.Gui.lambda$runModel$16(Gui.java:273)
	at java.lang.Thread.run(Thread.java:750)

It seems to run fine on smaller images.

Cheers
Pradeep

Reproduce the failure with the reported 4856 x 3672 image, then start at TileCalculator.java:254 and follow the calls from checkOutputSize and getOptimalTileSize. Check how BioimageIoModel.run reaches the tiling code. Done means large 2D images no longer fail with the reported NullPointerException.