github-actions[bot] opened a new issue, #33473:
URL: https://github.com/apache/beam/issues/33473
Performance change found in the
test:
`pytorch_image_classification_benchmarks-resnet152-GPU-mean_load_model_latency_milli_secs`
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).
`Test description:` Pytorch image classification on 50k images of size 224 x
224 with resnet 152 with Tesla T4 GPU.
Test link -
https://github.com/apache/beam/blob/42d0a6e3564d8b9c5d912428a6de18fb22a13ac1/.test-infra/jenkins/job_InferenceBenchmarkTests_Python.groovy#L151
Test dashboard -
http://metrics.beam.apache.org/d/ZpS8Uf44z/python-ml-runinference-benchmarks?orgId=1&viewPanel=7
```
timestamp: Tue Dec 31 06:55:07 2024, metric_value: 122599.20
timestamp: Mon Dec 30 06:45:37 2024, metric_value: 121750.85
timestamp: Sun Dec 29 06:48:51 2024, metric_value: 126396.92
timestamp: Sat Dec 28 06:44:31 2024, metric_value: 125432.50
timestamp: Fri Dec 27 06:51:24 2024, metric_value: 121294.73
timestamp: Thu Dec 26 06:46:05 2024, metric_value: 129072.67
timestamp: Wed Dec 25 06:41:52 2024, metric_value: 123215.38 <---- Anomaly
timestamp: Tue Dec 24 06:47:03 2024, metric_value: 131301.53
timestamp: Mon Dec 23 06:41:45 2024, metric_value: 135701.47
timestamp: Sun Dec 22 06:47:14 2024, metric_value: 124859.53
timestamp: Sat Dec 21 06:49:36 2024, metric_value: 130251.48
timestamp: Fri Dec 20 07:04:18 2024, metric_value: 132378.50
timestamp: Thu Dec 19 06:45:36 2024, metric_value: 130731.92
timestamp: Wed Dec 18 07:03:29 2024, metric_value: 127310.20
timestamp: Tue Dec 17 06:55:13 2024, metric_value: 128287.22
timestamp: Mon Dec 16 06:52:53 2024, metric_value: 129442.85
timestamp: Sun Dec 15 07:04:53 2024, metric_value: 138440.13
```
--
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]
