I’m trying to instantiate “intfloat/multilingual-e5-large” in jvm.
To do this, I’ll use this library:
<dependency>
<groupId>com.microsoft.onnxruntime</groupId>
<artifactId>onnxruntime</artifactId>
<version>1.15.1</version>
This model requires tokenizer and contains configuration for this.
onnxruntime
does not contain tools for such things. I only know about transformers.tokenization_auto
here. Maybe there’s some sort of java version of this library, or some alternatives?