How to Reproduce Previously Tracked Experiments

1 min
Kamil Kaczmarek
21st April, 2022

What will you learn?

How you can reproduce previously tracked experiments in Neptune.

  • 00:49

    Fetching the run data in Neptune

  • 01:33

    Customizing the runs table

  • 02:33

    Resuming the run in Neptune

  • 03:54

    Assigning the same data version to the run

  • 04:41

    Downloading the hyperparameters from the run

  • 07:12

    What will the results look like in the Neptune UI?

  • 07:59

    What is the logic for waiting (run.sync()) after doing some commands?

  • 10:06

    How to monitor the model through different phases of the MLOps lifecycle?

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

See also how to resume a run.

Check the docs about querying and downloading metadata. 

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.