explodinggradients / ragas

Supercharge Your LLM Application Evaluations 🚀
https://docs.ragas.io
Apache License 2.0
7.33k stars 747 forks source link

Exception raised in Job[300]: ValueError({'code': '3001', 'message': 'HTTP外部调用超时'}) #1600

Open snailmiss opened 3 weeks ago

snailmiss commented 3 weeks ago

how can i check and fix this valueError and TimeoutError? Evaluating: 77% 476/616 [09:54<‬01:58, 1.18it/s]2024-10-30 10:48:00.000 | INFO | Evaluating: 77% 477/616 [09:57<‬03:09, 1.37s/it]2024-10-30 10:48:01.612 | INFO | Evaluating: 79% 484/616 [10:05<‬03:00, 1.36s/it]2024-10-30 10:48:11.612 | INFO | Evaluating: 79% 486/616 [10:08<‬02:53, 1.34s/it]Exception raised in Job[391]: ValueError({'code': '3001', 'message': 'HTTP外部调用超时'}) Evaluating: 80% 490/616 [10:14<‬03:06, 1.48s/it]2024-10-30 10:48:21.612 | INFO | Evaluating: 80% 491/616 [10:18<‬04:33, 2.19s/it]Prompt fix_output_format failed to parse output: The output parser failed to parse the output after 0 retries. Prompt fix_output_format failed to parse output: The output parser failed to parse the output after 0 retries. Prompt correctness_classifier failed to parse output: The output parser failed to parse the output after 0 retries. Exception raised in Job[76]: RagasOutputParserException(The output parser failed to parse the output after 0 retries.) Evaluating: 80% 493/616 [10:19<‬03:01, 1.48s/it]Exception raised in Job[132]: TimeoutError()

jjmachan commented 3 weeks ago

hey @snailmiss which model are you using for evaluation?

you can also checkout https://docs.ragas.io/en/stable/howtos/customizations/_run_config/ for how to configure timeouts

snailmiss commented 3 weeks ago

you can also checkout https://docs.ragas.io/en/stable/howtos/customizations/_run_config/ for how to configure timeouts

@jjmachan how to set maxworks and timeout,i set maxworks=64,timeout=180,still have ValueError and TimeoutError。

jjmachan commented 2 weeks ago

@snailmiss maybe you could try batch parameter for evaluate() function too? image