RkGrit opened a new pull request, #16850: URL: https://github.com/apache/iotdb/pull/16850
The InferenceRequestPool subprocess no longer loads the model through the ModelManager instance, but directly calls the load_model method, and the parameter passed to it is model_info instead of model_storage. In this way, ModelManager and ModelStorage are only controlled by the main process, but it is not responsible for loading the model. During inference, the InferenceRequestPool subprocess loads the model as needed through the Pipeline. Final effect: Main process manages models, child process loads models. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
