Organizing and filtering Runs
You can find, organize, sort, and filter all the Runs from a project in the Runs space of the Neptune UI.
You will learn about:
Customizing Columns: Add and customize metrics, parameters, dates, and other metadata in columns
Searching and filtering Runs: Filter Runs by the value of a metric, parameter, date, and other metadata
Saving Runs table Views: Save views of the Runs table to be used by any member of the project later
You can configure what metadata logged to Neptune is displayed as columns in the Runs table.
You can see:
metrics,
parameters,
text logs,
system parameters.
and any other metadata
Click +Add column button
Type in a name of the metric or parameter you want to add and click on it
Choose a color of the column, notation format, and a name if you want to change the default.
Drag your column wherever you want it to be
Note You can use suggested Learn more how to log different types of meta-data: What objects can you log to Neptune.
Neptune UI automatically proposes columns based on what is different between Runs.
Suggested columns are the right-most columns in the dashboard.
You can:
add a suggested column by clicking on a + icon
get a new column suggestion by clicking on a - icon
For Fields where you logged a series of values, you can choose an aggregate function and add it to the Runs table.
This is useful for training losses where you want to show the best (min or max) value of a loss during training.
You can display:
MIN: minimal value
MAX: maximal value
AVERAGE: average or mean value
LAST: the value that was logged last
VARIANCE: variance calculated on the logged values
You can search and filter Runs in Neptune based on any Field.
To filter Runs, start typing the Fields' name into the filter box and select the field you are looking for. Select the most fitting operator and provide reference value if required by the chosen operator.
Note You can enter queries using the keyboard only. Use up and down arrows to move around and Enter to select and accept.
Each field type has a set of specific operators that lets you specify what you are looking for.
You can:
use both the full path of the Field and a custom column name to search columns
combine many filters together to narrow the search as needed
Examples:
Find models small enough to be used on mobile (
model/size <= 50 MB) that have a decent accuracy (test/acc > 0.90)Check all runs from this month (
sys/creation_time > YYYY.MM.DD) that failed during training (sys/failed = True)Find all my runs (
sys/owner = "giulia") from my current task (sys/tags ∋ "exploration")
You can access all your previous search queries with the Recent searches.
You can sort the Runs table by any column including aggregate columns like max or last value of series of training metrics.
To do that, Click on the arrowhead symbol next to the column name.
Once you have:
selected and organized the columns how you like them
filtered Runs based on metrics and parameters
You may want to save the Runs table view for later. It will be visible to any member of your project.
Neptune lets you save one or many views of the Runs table so that you can:
Look at different parameters and metrics for a particular idea or experiment
Have your teammates look at different columns and save them without breaking your views
Switch between different aspects of your experiments and model training runs quickly
See example Runs table view in Neptune
In this short tutorial, you will learn:
how to save the column configuration and filters you have set up to a new table view
how to pin this view as the default for the project
Important To save table views, you need to be the "contributor" or "owner" of the project. For more information, see Roles in project.
To create a new Runs table view you need to:
click on Save as new button. A model will open.
give your new table view a name. You can change it later
find your table view in a list of views and pin it as default by clicking the pin icon
Now you and every other project member can:
find and select this view and see parameters, metrics, and filters you configured
share this view by sending a link
