How to Explore a Single Run or Experiment

1 min
Kamil Kaczmarek
21st April, 2022

What will you learn?

How you can log and explore time-series forecasting runs/experiments in Neptune.

  • 00:12

    How to add Neptune to your project?

  • 00:24

    How to connect to Neptune and create a run?

  • 01:05

    Logging metadata in a hierarchical structure

  • 01:55

    Uploading the matplotlib figures as images

  • 03:11

    Logging different scores

  • 03:20

    Uploading the model checkpoints

  • 03:49

    How can I visualize all the metadata logged?

  • 05:16

    How does Neptune collect the Git information?

  • 05:42

    Can I stop Neptune from logging my Git info?

  • 05:50

    How to visualize and inspect the training metrics?

  • 06:51

    How to quickly make sense of the most relevant run’s metadata?

  • 06:57

    Exploring the run-level custom dashboards

  • 08:13

    Are the custom dashboards available to all runs in a project?

Important: This video was created in February 2022. For the most up-to-date code examples, please refer to the Neptune docs

Other useful resources

Check also the docs on what you can log and display in Neptune. 

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.