NeptuneResultTracker
- class NeptuneResultTracker(project_qualified_name: str | None = None, api_token: str | None = None, offline: bool = False, experiment_id: int | None = None, experiment_name: str | None = None, tags: Collection[str] | None = None)[source]
Bases:
ResultTracker
A tracker for Neptune.ai.
Initialize the Neptune result tracker.
- Parameters:
project_qualified_name (str | None) – Qualified name of a project in a form of
namespace/project_name
. IfNone
, the value ofNEPTUNE_PROJECT
environment variable will be taken. For testing, should be <your username>/sandboxapi_token (str | None) –
User’s API token. If
None
, the value ofNEPTUNE_API_TOKEN
environment variable will be taken.Note
It is strongly recommended to use
NEPTUNE_API_TOKEN
environment variable rather than placing your API token in plain text in your source code.offline (bool) – Run neptune in offline mode (uses
neptune.OfflineBackend
as the backend)experiment_id (int | None) – The identifier of a pre-existing experiment to use. If not given, will rely on the
experiment_name
.experiment_name (str | None) – The name of the experiment. If no
experiment_id
is given, one will be created based on the name.tags (Collection[str] | None) – A collection of tags to add to the experiment
- Raises:
ValueError – If neither an experiment name nor experiment ID is given
Methods Summary
log_metrics
(metrics[, step, prefix])Log metrics to result store.
log_params
(params[, prefix])Log parameters to result store.
Methods Documentation