NeptuneResultTracker¶
-
class
NeptuneResultTracker
(project_qualified_name=None, api_token=None, offline=False, experiment_id=None, experiment_name=None, tags=None)[source]¶ Bases:
pykeen.trackers.base.ResultTracker
A tracker for Neptune.ai.
Initialize the Neptune result tracker.
- Parameters
project_qualified_name (
Optional
[str
]) – Qualified name of a project in a form ofnamespace/project_name
. IfNone
, the value ofNEPTUNE_PROJECT
environment variable will be taken. For testing, should be <your username>/sandboxUser’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 (usesneptune.OfflineBackend
as the backend)experiment_id (
Optional
[int
]) – The identifier of a pre-existing experiment to use. If not given, will rely on theexperiment_name
.experiment_name (
Optional
[str
]) – The name of the experiment. If noexperiment_id
is given, one will be created based on the name.tags (
Optional
[Collection
[str
]]) – A collection of tags to add to the experiment
Methods Summary
log_metrics
(metrics[, step, prefix])Log metrics to result store.
log_params
(params[, prefix])Log parameters to result store.
Methods Documentation