AnandInguva opened a new issue, #24849:
URL: https://github.com/apache/beam/issues/24849
Performance change found in the
test:
`apache_beam.testing.benchmarks.inference.pytorch_image_classification_benchmarks_22`
for the metric: `mean_load_model_latency_milli_secs`.
For more information on how to triage the alerts, please look at
`Triage performance alert issues` section of the
[README](https://github.com/apache/beam/tree/master/sdks/python/apache_beam/testing/analyzers/README.md#triage-performance-alert-issues).
timestamp: Mon Dec 26 16:54:29 2022, metric_value: `94862.48666666666`
timestamp: Sun Dec 25 16:51:18 2022, metric_value: `98225.24666666667`
timestamp: Sat Dec 24 16:51:59 2022, metric_value: `95972.68666666666`
timestamp: Fri Dec 23 16:55:13 2022, metric_value: `99635.41333333333`
timestamp: Thu Dec 22 16:50:27 2022, metric_value: `98171.96`
timestamp: Wed Dec 21 16:53:28 2022, metric_value: `97866.29333333333`
timestamp: Tue Dec 20 16:55:41 2022, metric_value: `98970.39333333333`
timestamp: Mon Dec 19 16:56:58 2022, metric_value: `120506.56666666667`
timestamp: Sun Dec 18 16:54:16 2022, metric_value: `97121.59333333334`
timestamp: Sat Dec 17 16:54:01 2022, metric_value: `97628.54`
timestamp: Fri Dec 16 16:56:51 2022, metric_value: `96672.42666666667` <----
Anomaly
timestamp: Thu Dec 15 16:53:11 2022, metric_value: `104591.26`
timestamp: Wed Dec 14 16:56:42 2022, metric_value: `108879.96`
timestamp: Tue Dec 13 16:53:29 2022, metric_value: `116157.17333333334`
timestamp: Mon Dec 12 16:55:39 2022, metric_value: `107796.69333333333`
timestamp: Sun Dec 11 16:54:38 2022, metric_value: `111908.9`
timestamp: Sat Dec 10 16:58:10 2022, metric_value: `107900.74`
timestamp: Fri Dec 9 16:56:08 2022, metric_value: `106021.94`
timestamp: Thu Dec 8 16:55:37 2022, metric_value: `107807.34666666666`
timestamp: Wed Dec 7 16:57:05 2022, metric_value: `106788.94`
timestamp: Tue Dec 6 16:55:53 2022, metric_value: `112162.67333333334`
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
