[ https://issues.apache.org/jira/browse/OPENNLP-1515?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17774502#comment-17774502 ]
ASF GitHub Bot commented on OPENNLP-1515: ----------------------------------------- kinow commented on PR #551: URL: https://github.com/apache/opennlp/pull/551#issuecomment-1759586922 > What about using `<profiles>` and `<os>`-based activation via Maven? I think that would be for build only? Unless we produced artefacts to be uploaded to Maven per OS (I think I saw that once, some different flag in the dependency in Maven pom.xml, but never actually used it?) > Default to onnxruntime instead of onnxruntime-gpu > ------------------------------------------------- > > Key: OPENNLP-1515 > URL: https://issues.apache.org/jira/browse/OPENNLP-1515 > Project: OpenNLP > Issue Type: Task > Components: Deep Learning > Reporter: Jeff Zemerick > Assignee: Jeff Zemerick > Priority: Major > > The onnxruntime-gpu dependency is currently being included by opennlp-dl. > <dependency> > <groupId>com.microsoft.onnxruntime</groupId> > <!-- This dependency supports CPU and GPU --> > <artifactId>onnxruntime_gpu</artifactId> > <version>${onnxruntime.version}</version> > </dependency> > The problem is, GPU support is only on Linux and Windows and not OSX. I think > it would be best to use the onnxruntime dependency instead. > But we need to make OpenNLP able to use GPU easily. -- This message was sent by Atlassian Jira (v8.20.10#820010)