Matt Camp

Pod goes OutOfnvidia.com/gpu before k8s-device-plugin is ready

5

I have an issue when using cluster autoscaling for GPU nodes. I am using Karpenter as the cluster autoscaler and I'm trying to deploy NVidia Riva. The pod deployment spec...

This PR adds the ability to push training and evaluation metrics to InfluxDB (via Telegraf). When combined with https://github.com/aws-deepracer-community/deepracer-for-cloud/pull/159 it should allow for some nice interactive Grafana dashboards. As the...

Add Telegraf/InfluxDB/Grafana compose stack for recording InfluxDB metrics

1

This PR adds a docker-compose stack which launches three additional services - Telegraf to accept UDP push metrics and pass to InfluxDB - InfluxDB to store time-series metrics - Grafana...

feat: Add feature to store custom_files and config in a subdir for each experiment

2

The feature enables each training session to have it's config and custom_files stored within a subdir under `experiments/`. This simplifies being able to locate the config and files that were...