Could not initialize class ai.djl.onnxruntime.engine.0rtNDManager about djl HOT 4 CLOSED

When you usine on K8s, are you using DJLServing? We have kserve integration for DJLServing.

If you set ai.djl.onnx.disable_alternative, you should not use use OnnxRuntime engine to do any NDArray operations for pre/post processing. See: https://docs.djl.ai/docs/hybrid_engine.html

Can you provide more detail about use case, configuration and any code that can reproduce your error?

from djl.

I found that the other way doesn't report this error, when k8s starts the service, if the network is open, it will download the following:
2024-02-20 17:50:31.784 INFO 7 --- [9:8816-thread-2] ai.djl.pytorch.jni.LibUtils : Downloading https://publish.djl.ai/pytorch/1.13.1/cpu/linux-x86_64/native/lib/libc10.so.gz ...
2024-02-20 17:50:33.111 INFO 7 --- [9:8816-thread-2] ai.djl.pytorch.jni.LibUtils : Downloading https://publish.djl.ai/pytorch/1.13.1/cpu/linux-x86_64/native/lib/libtorch_cpu.so.gz ...
2024-02-20 17:51:52.903 INFO 7 --- [9:8816-thread-2] ai.djl.pytorch.jni.LibUtils : Downloading https://publish.djl.ai/pytorch/1.13.1/cpu/linux-x86_64/native/lib/libtorch.so.gz ...
2024-02-20 17:51:53.157 INFO 7 --- [9:8816-thread-2] ai.djl.pytorch.jni.LibUtils : Downloading https://publish.djl.ai/pytorch/1.13.1/cpu/linux-x86_64/native/lib/libgomp-52f2fd74.so.1.gz ...
2024-02-20 17:51:53.871 INFO 7 --- [9:8816-thread-2] ai.djl.pytorch.jni.LibUtils : Downloading jni https://publish.djl.ai/pytorch/1.13.1/jnilib/0.23.0/linux-x86_64/cpu/libdjl_torch.so to cache ...
. But I'm in a network-free environment right now, so it will occurred this error:
java.lang.NoClassDefFoundError: Could not initialize class ai.djl.onnxruntime.engine.OrtNDManager
and I want to put these LibUtils packages into the jar package, Or do you have a better way?

When you usine on K8s, are you using DJLServing? We have kserve integration for DJLServing.

If you set ai.djl.onnx.disable_alternative, you should not use use OnnxRuntime engine to do any NDArray operations for pre/post processing. See: https://docs.djl.ai/docs/hybrid_engine.html

Can you provide more detail about use case, configuration and any code that can reproduce your error?

from djl.

You need use offline PyTorch library in your project. See: https://docs.djl.ai/engines/pytorch/pytorch-engine/index.html#linux-cpu

You can also consider create a fatjar, you can follow this example: https://github.com/deepjavalibrary/djl-demo/tree/master/development/fatjar#how-to-distribute-djl-application

from djl.

ok thank you ，i will try after

from djl.