Resources

How to Log, Explore, Compare the RL Agent Training Metadata

0 min
Kamil Kaczmarek
21st April, 2022

What will you learn?

How you can log, explore and compare the RL agent training metadata in Neptune.

  • 00:12

    Introduction to the CartPole RL problem

  • 00:40

    How to connect to Neptune and create a run?

  • 01:18

    Logging metadata in a hierarchical structure

  • 02:15

    How to log charts, metrics, agent-specific metadata to Neptune?

  • 03:41

    How to log images and other artifacts to Neptune?

  • 04:16

    Logging a series of images

  • 04:47

    Uploading the model

  • 05:04

    What will the results look like in the Neptune UI?

  • 06:26

    Some visualizations of the input screen and the training episode

  • 07:08

    Exploring the dashboards

Other useful resources

See also the article to learn how to make sense of the reinforcement learning agents.Ā 

Couldn’t find the use case you were looking for?

Just get in touch, and our ML team will create a custom demo for you.