environments

package
v0.3.0 Latest Latest
Warning

This package is not in the latest version of its module.

Go to latest
Published: May 18, 2026 License: MIT Imports: 3 Imported by: 0

Documentation

Index

Constants

This section is empty.

Variables

This section is empty.

Functions

func SaveTrajectory

func SaveTrajectory(traj *Trajectory, path string) error

SaveTrajectory 保存轨迹

Types

type Environment

type Environment interface {
	Reset() (map[string]interface{}, error)
	Step(action map[string]interface{}) (map[string]interface{}, error)
	Observation() map[string]interface{}
	Close() error
}

Environment RL 训练环境接口

type ToolEnv

type ToolEnv struct {
	// contains filtered or unexported fields
}

ToolEnv 基于工具的RL环境

func NewToolEnv

func NewToolEnv(tools []string) *ToolEnv

NewToolEnv 创建工具使用环境

func (*ToolEnv) Close

func (e *ToolEnv) Close() error

func (*ToolEnv) Observation

func (e *ToolEnv) Observation() map[string]interface{}

func (*ToolEnv) Reset

func (e *ToolEnv) Reset() (map[string]interface{}, error)

func (*ToolEnv) Step

func (e *ToolEnv) Step(action map[string]interface{}) (map[string]interface{}, error)

type Trajectory

type Trajectory struct {
	TaskID      string           `json:"task_id"`
	Steps       []TrajectoryStep `json:"steps"`
	FinalReward float64          `json:"final_reward"`
}

Trajectory 轨迹

type TrajectoryStep

type TrajectoryStep struct {
	Observation map[string]interface{} `json:"observation"`
	Action      map[string]interface{} `json:"action"`
	Reward      float64                `json:"reward"`
}

TrajectoryStep 轨迹步骤

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL