Performance Regression or Improvement: pytorch_image_classification_benchmarks-resnet101-mean_load_model_latency_milli_secs:mean_load_model_latency_milli_secs

[github-actions]

Performance change found in the
test: pytorch_image_classification_benchmarks-resnet101-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.

Test description: Pytorch image classification on 50k images of size 224 x 224 with resnet 101.
Test link -

beam/.test-infra/jenkins/job_InferenceBenchmarkTests_Python.groovy

Line 34 in 42d0a6e

test : 'apache_beam.testing.benchmarks.inference.pytorch_image_classification_benchmarks',

Test dashboard - http://metrics.beam.apache.org/d/ZpS8Uf44z/python-ml-runinference-benchmarks?orgId=1&viewPanel=7

timestamp: Sun Oct 29 18:23:28 2023, metric_value: 65695.24
timestamp: Sun Oct 29 10:18:41 2023, metric_value: 65351.07
timestamp: Sun Oct 29 04:30:42 2023, metric_value: 65757.73
timestamp: Sun Oct 29 04:23:41 2023, metric_value: 67388.44
timestamp: Sat Oct 28 18:23:29 2023, metric_value: 67372.80
timestamp: Sat Oct 28 10:00:26 2023, metric_value: 64035.43
timestamp: Sat Oct 28 04:30:54 2023, metric_value: 65648.27
timestamp: Sat Oct 28 04:18:53 2023, metric_value: 67149.77 <---- Anomaly
timestamp: Fri Oct 27 18:26:05 2023, metric_value: 103494.35
timestamp: Fri Oct 27 04:31:22 2023, metric_value: 121780.65
timestamp: Fri Oct 27 04:24:27 2023, metric_value: 73791.83
timestamp: Thu Oct 26 18:25:08 2023, metric_value: 77371.03
timestamp: Thu Oct 26 14:24:44 2023, metric_value: 75632.53
timestamp: Thu Oct 26 04:32:06 2023, metric_value: 73552.71
timestamp: Thu Oct 26 04:18:45 2023, metric_value: 73064.23
timestamp: Wed Oct 25 18:23:51 2023, metric_value: 81966.59
timestamp: Wed Oct 25 10:24:17 2023, metric_value: 70373.45
timestamp: Wed Oct 25 04:29:48 2023, metric_value: 74224.48

[AnandInguva]

https://screenshot.googleplex.com/6fmMSinSjvBpyyU

There are couple of spikes and later it went down. Now since GHA migration, the job seems to be running 4 times a day, which i not needed for Performance tests. I will put a PR to run the job once per day.