search

deepjavalibrary / djl

An Engine-Agnostic Deep Learning Framework in Java
https://djl.ai
Apache License 2.0
4.05k stars 648 forks source link

Cross compatibility between Pytorch native versions and Pytorch engines/apis #544

Closed carlosuc3m closed 3 years ago

carlosuc3m commented 3 years ago

Question

Hello there, I was wondering if there exists a document secifying the cross compatibility between Pytorch native versions and Pytorch engines/apis. I have not been able to found anything and in order to find it out I have been doing it by trial and error. Thank you for your time, Carlos

zachgk commented 3 years ago

You can take a look at the PyTorch engine README document: https://github.com/awslabs/djl/blob/master/pytorch/pytorch-engine/README.md. There is only a single PyTorch version that matches each version of DJL. If you view this page at a version tag rather than the master branch, you can see what versions are supported by the various released versions of DJL.

frankfliu commented 3 years ago

@carlosuc3m The recommended way is to use DJL's bom to manage your dependencies: http://docs.djl.ai/master/bom/index.html You can find the bom pom.xml file from maven central for each version: https://search.maven.org/artifact/ai.djl/bom/0.9.0/pom

lanking520 commented 3 years ago

@carlosuc3m Close this issue for now. If you have any more concerns about the versioning issue, please feel free to reopen!