data_juicer.ops.mapper.agent_tool_relevance_mapper module#
LLM: tool/capability fit vs user task (1–5).
Inspired by OpenJudge tool-selection; DJ: meta tags or text-only, no schema.
Reference: https://agentscope-ai.github.io/OpenJudge/built_in_graders/agent_graders/ #toolselectiongrader
- class data_juicer.ops.mapper.agent_tool_relevance_mapper.AgentToolRelevanceMapper(*args, **kwargs)[源代码]#
基类:
_DialogQualityLLMMapperBaseRough fit between tools/capabilities and the user task (uses meta tool tags).
- OP_NAME = 'agent_tool_relevance_mapper'#
- META_KEY = 'agent_tool_relevance'#
- EVAL_KIND = 'agent_tool'#