caution

This function is experimental and might be changed or removed in a future minor release. Use with caution in production code.

fetch_experiments_table_global()

Python package: neptune-query

Fetches a table of experiment metadata, with runs as rows and attributes as columns. The query targets projects globally, across all workspaces that the user can access with the provided API token.

To narrow the results, define filters for experiments to search or attributes to include.

Parameters

experiments

str | list of str | Filter

optional

default: None

Filter specifying which experiments to include.

• If a string is provided, it's treated as a regex pattern that the names must match.
• If a list of strings is provided, it's treated as exact experiment names to match.
• To provide a more complex condition on an arbitrary attribute value, pass a Filter object:
  • The filter must use only select attributes from the sys and env namespaces. Ask your administrator for a list of supported attributes.
  • Use the Attribute class to specify the attribute name and type.

attributes

str | list of str | AttributeFilter

optional

default: None

Filter specifying which attributes to include.

• If a string is provided, it's treated as a regex pattern that the attribute names must match.
• If a list of strings is provided, it's treated as exact attribute names to match.
• To provide a more complex condition, pass an AttributeFilter object.

sort_by

Attribute

optional

default: None

Attribute to sort the table by. If provided, needs to specify the attribute type.

sort_direction

"asc" | "desc"

optional

default: "desc"

The direction to sort columns by: "desc" (default) or "asc".

limit

int

optional

default: None

Maximum number of experiments to return. By default, all experiments are included.

type_suffix_in_column_names

bool

optional

default: False

If True, columns of the returned DataFrame are suffixed with :<type>. For example, "attribute1:float_series", "attribute1:string".

If False (default), the method throws an exception if there are multiple types under one path.

Returns

pandas.DataFrame – A DataFrame similar to the runs table in the web app.

The DataFrame has:

• a single-level index "experiment" with experiment names
• a single-level column index with attribute names

For series attributes, the last logged value is returned.

Raises

• AttributeTypeInferenceError – If the attribute type wasn't specified in a filter passed to the experiments argument.
• ConflictingAttributeTypes – If there are conflicting attribute types under the same path and the type_suffix_in_column_names argument is set to False.

Example

Fetch attributes matching loss or configs from two specific experiments across all projects:

import neptune_query.experimental as nq_experimental


nq_experimental.fetch_experiments_table_global(
    experiments=["seagull-week1", "seagull-week2"],
    attributes=r"loss | configs",
)

Fetch attributes matching loss or configs from experiments created by a specific user across all projects:

import neptune_query.experimental as nq_experimental
from neptune_query.filters import Filter, Attribute


nq_experimental.fetch_experiments_table_global(
    experiments=Filter.eq(Attribute("sys/owner", "string"), "my-username"),
    attributes=r"loss | configs",
)