evaluation

package

v1.10.3 Latest Latest Go to latest Published: Dec 1, 2025 License: Apache-2.0 Imports: 14 Imported by: 0

Details

This section is empty.

This section is empty.

func Evaluate(ctx context.Context, out Printer, agentFilename, evalsDir string, runConfig *config.RuntimeConfig) error

func Save(sess *session.Session) (string, error)

type Printer interface {
	Printf(format string, a ...any)
}

type Result struct {
	EvalFile     string
	FirstMessage string
	Score        Score
	Cost         float64
	OutputTokens int64
}

type Score struct {
	ToolTrajectoryScore float64
	Rouge1Score         float64
}