Package org.apache.beam.sdk.ml.inference.remote

Class RemoteInference

java.lang.Object

org.apache.beam.sdk.ml.inference.remote.RemoteInference

public class RemoteInference
extends java.lang.Object

A PTransform for making remote inference calls to external machine learning services.

RemoteInference provides a framework for integrating remote ML model inference into Apache Beam pipelines and handles the communication between pipelines and external inference APIs.

Example: OpenAI Model Inference

 // Create model parameters
 OpenAIModelParameters params = OpenAIModelParameters.builder()
     .apiKey("your-api-key")
     .modelName("gpt-4")
     .instructionPrompt("Analyse sentiment as positive or negative")
     .build();

 // Apply remote inference transform
 PCollection<OpenAIModelInput> inputs = pipeline.apply(Create.of(
     OpenAIModelInput.create("An excellent B2B SaaS solution that streamlines business processes efficiently."),
     OpenAIModelInput.create("Really impressed with the innovative features!")
 ));

 PCollection<Iterable<PredictionResult<OpenAIModelInput, OpenAIModelResponse>>> results =
     inputs.apply(
         RemoteInference.<OpenAIModelInput, OpenAIModelResponse>invoke()
             .handler(OpenAIModelHandler.class)
             .withParameters(params)
     );
 

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	RemoteInference.Invoke<InputT extends BaseInput,OutputT extends BaseResponse>

Method Summary

Modifier and Type	Method	Description
static <InputT extends BaseInput,OutputT extends BaseResponse> RemoteInference.Invoke<InputT,OutputT>	invoke()	Invoke the model handler with model parameters.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

invoke

public static <InputT extends BaseInput,OutputT extends BaseResponse> RemoteInference.Invoke<InputT,OutputT> invoke()

Invoke the model handler with model parameters.