GPU trained model compile error

General
,
1

Hello

Yolov8n object detectoin model trained cutom dataset by GPU(RTX 4070)

But that model when trying hailomz compile, logs here

hailo_model_optimization.acceleras.utils.acceleras_exceptions.NegativeSlopeExponentNonFixable: Quantization failed in layer yolov8n/conv63 due to unsupported required slope. Desired shift is 10.0, but op has only 8 data bits. This error raises when the data or weight range are not balanced. Mostly happens when using random calibration-set/weights, the calibration-set is not normalized properly or batch-normalization was not used during training.

bash: --hw-arch=hailo8l: command not found

(When i try CPU trained, it work! any parameter didn’t change)

Hailo didn’t support GPU trained model?

if not, how to compile GPU trained model?

And, this is another question what is the image limit size? hailo supported any size image?