Skip to main content
App version: 3.20250811
caution

You're viewing docs for the deprecated Fetcher API.

We recommend upgrading to the new Query API.

Attribute

Python package: neptune-fetcher

Helper for specifying an attribute and picking a metric aggregation function.

When fetching experiments or runs, use the Attribute constructor to filter and sort the returned entries.

Related

Parameters​

name
str
required

An attribute name to match exactly.

aggregation
"last" | "min" | "max" | "average" | "variance"
optional
default: None

Aggregation function to apply when specifying a metric of type FloatSeries.

Defaults to "last", that is, the last logged value.

type
"float" | "int" | "string" | "bool" | "datetime" | "float_series" | "string_set" | "string_series"
optional
default: None

Attribute type. Specify it to resolve ambiguity, in case some of the project's runs contain attributes that have the same name but are of a different type.

Examples​

Select a metric and pick variance as the aggregation:

import neptune_fetcher.alpha as npt
from neptune_fetcher.alpha.filters import Attribute, Filter


val_loss_variance = Attribute(
    name="val/loss",
    aggregation="variance",
)

Construct a filter around the attribute with the Filter class, then pass it to a fetching or listing method:

tiny_val_loss_variance = Filter.lt(val_loss_variance, 0.01)
npt.fetch_experiments_table(experiments=tiny_val_loss_variance)