* 批量评估 * @param agentId Agent ID * @param testCases 测试用例列表 * @param scorers Scorer 配置列表 * @param concurrency 并发数 * @returns Eval 结果
(agentId: string, testCases: TestCase[], scorers: ScorerConfig[], concurrency?: number)
| 387 | * @returns Eval 结果 |
| 388 | */ |
| 389 | async batchEval(agentId: string, testCases: TestCase[], scorers: ScorerConfig[], concurrency?: number): Promise<EvalResult> { |
| 390 | const evalInfo = await this.create({ |
| 391 | name: `Batch Eval - ${new Date().toISOString()}`, |
| 392 | type: "batch", |
| 393 | agentId, |
| 394 | testCases, |
| 395 | scorers, |
| 396 | concurrency, |
| 397 | }); |
| 398 | |
| 399 | return this.waitForCompletion(evalInfo.id); |
| 400 | } |
| 401 | |
| 402 | /** |
| 403 | * 比较两个 Agents |
no test coverage detected