github-actions[bot] opened a new issue, #39375:
URL: https://github.com/apache/beam/issues/39375

   
     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: Fri Jul 17 10:18:22 2026, metric_value: 126438.93
   timestamp: Thu Jul 16 11:02:44 2026, metric_value: 127597.95
   timestamp: Wed Jul 15 11:18:41 2026, metric_value: 132462.05
   timestamp: Tue Jul 14 11:15:32 2026, metric_value: 136065.18 <---- Anomaly
   timestamp: Mon Jul 13 10:48:37 2026, metric_value: 120508.23
   timestamp: Sun Jul 12 10:25:31 2026, metric_value: 122502.82
   timestamp: Sat Jul 11 10:10:43 2026, metric_value: 117655.37
   timestamp: Fri Jul 10 10:58:22 2026, metric_value: 119218.48
   timestamp: Thu Jul  9 10:50:40 2026, metric_value: 123824.00
   timestamp: Wed Jul  8 10:49:09 2026, metric_value: 123250.77
   timestamp: Wed Jul  8 02:43:33 2026, metric_value: 123558.67
   timestamp: Tue Jul  7 11:18:55 2026, metric_value: 133004.28
   timestamp: Mon Jul  6 11:19:13 2026, metric_value: 127438.88
   timestamp: Sun Jul  5 11:06:56 2026, metric_value: 119981.35
   
   ```


-- 
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]

Reply via email to