Make the metric call function async

[ZiTao-Li]

AgentScope Version

1.0.2

Description

Change the __call__ function of metric to async function for more general usage (e.g., LLM as a judge metrics)

• update related modules
• update tutorial
• add a test for the evaluators in evaluate module

Checklist

Please check the following items before code is ready to be reviewed.

• Code has been formatted with pre-commit run --all-files command
• All tests are passing
• Docstrings are in Google style
• Related documentation has been updated (e.g. links, examples, etc.)
• Code is ready for review

[DavdGao]

Please see inline comments, others look good to me.

[Copilot]

Pull Request Overview

This PR changes the metric __call__ function from synchronous to asynchronous to support more general usage patterns, particularly LLM-as-a-judge metrics that require async operations.

Key changes:

• Made the MetricBase.__call__ method abstract and async
• Updated all evaluator classes to handle async metric calls
• Added comprehensive test coverage for the evaluation module

Reviewed Changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
tests/evaluation_test.py	Added comprehensive test suite for evaluators with async metric support
src/agentscope/evaluate/_task.py	Made task evaluation method async to support async metric calls
src/agentscope/evaluate/_metric_base.py	Changed abstract metric __call__ method to async
src/agentscope/evaluate/_evaluator/_ray_evaluator.py	Refactored Ray evaluator to use async actors and proper async handling
src/agentscope/evaluate/_evaluator/_general_evaluator.py	Updated general evaluator to handle async metric evaluation
src/agentscope/evaluate/_ace_benchmark/_ace_metric.py	Made ACE benchmark metrics async-compatible
docs/tutorial/zh_CN/src/task_eval.py	Updated Chinese tutorial example with async metric and corrected name
docs/tutorial/en/src/task_eval.py	Updated English tutorial example with async metric and corrected name

Comments suppressed due to low confidence (1)

src/agentscope/evaluate/_evaluator/_ray_evaluator.py:1

• Using __file__ in py_modules for Ray runtime_env is incorrect. __file__ refers to the current Python file being executed, but py_modules expects module names or paths to modules that should be made available to Ray workers. This will likely cause import errors in Ray workers.

# -*- coding: utf-8 -*-

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[DavdGao]

LGTM