包package io.agentscope.core.training.strategy

io.agentscope.core.training.strategy

package io.agentscope.core.training.strategy

io.agentscope.core.training.strategy 属于评估与追踪，评测、Studio 与可观测性。该包共收录 7 个公开类型。

io.agentscope.core.training.strategy belongs to the Evaluation & tracing layer and contains 7 public types.

接口Interfaces

类型Type	种类Kind	说明Description
`TrainingSelectionStrategy`	接口interface	定义 Training Selection Strategy契约的接口。 Training Selection Strategy - Training request filtering strategy

类型Type	种类Kind	说明Description
`ExplicitMarkingStrategy`	类class	Explicit Marking Strategy，对智能体行为进行评测、追踪与可观测性分析。 Explicit Marking Strategy - Selection strategy based on user explicit marking
`SamplingRateStrategy`	类class	Sampling 速率 Strategy，对智能体行为进行评测、追踪与可观测性分析。 Sampling Rate Strategy - Selection strategy based on sampling rate
`SelectionDecision`	类class	Selection 决策，对智能体行为进行评测、追踪与可观测性分析。 Selection Decision - Selection decision result
`TrainingAnnotation`	类class	Training 注解，对智能体行为进行评测、追踪与可观测性分析。 TrainingAnnotation represents a training annotation for an agent call.
`TrainingContext`	类class	Training 上下文，对智能体行为进行评测、追踪与可观测性分析。 TrainingContext provides API for marking agent calls for training.