github-actions[bot] commented on issue #27986:
URL: https://github.com/apache/beam/issues/27986#issuecomment-1962071521
Performance change found in the
test:
`pytorch_image_classification_benchmarks-resnet152-GPU-mean_inference_batch_latency_micro_secs`
for the metric: `mean_inference_batch_latency_micro_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?from=now-90d&to=now&viewPanel=2
```
timestamp: Fri Feb 23 06:53:41 2024, metric_value: 4560157.85
timestamp: Wed Feb 21 07:12:25 2024, metric_value: 4883101.24
timestamp: Tue Feb 20 06:53:46 2024, metric_value: 5398728.80 <---- Anomaly
timestamp: Sun Feb 18 06:41:31 2024, metric_value: 3060535.64
timestamp: Sat Feb 17 06:52:20 2024, metric_value: 2565324.57
timestamp: Fri Feb 16 06:55:15 2024, metric_value: 3789397.17
timestamp: Thu Feb 15 06:58:53 2024, metric_value: 4431068.78
timestamp: Wed Feb 14 06:50:20 2024, metric_value: 3481961.73
timestamp: Mon Feb 12 06:42:22 2024, metric_value: 3489616.96
timestamp: Sat Feb 10 06:45:00 2024, metric_value: 3158261.25
timestamp: Fri Feb 9 06:47:02 2024, metric_value: 2655775.48
timestamp: Thu Feb 8 06:47:05 2024, metric_value: 2885917.85
timestamp: Wed Feb 7 06:52:04 2024, metric_value: 3010852.66
```
--
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]
