wandb.load_wandb_history

wandb.load_wandb_history(runs, *, keys=None, x='_step', api=None)

Read the metrics Weights & Biases runs logged.

Parameters

Name	Type	Description	Default
runs	str, os.PathLike, wandb.apis.public.Run, mapping, or iterable of these	The runs: a run’s file, `run-<id>.wandb`, or a folder holding run files, such as `wandb/`, every run under it read; a run’s path on the W&B server (`"entity/project/run_id"`, as its page’s URL ends); a run `wandb.Api().run()` or `.runs()` returned; or an iterable of these. A string that names a file or folder on disk is read from disk. A mapping instead reads history saved earlier: run name to its rows, as a list of dicts or as the `pandas.DataFrame` that `run.history()` returns.	required
keys	iterable of str	Only these metrics, in this order. By default every metric a run logged a number under, sorted, leaving out W&B’s own `_`-prefixed keys and `x`. A metric that no run logged is warned about.	`None`
x	str	The key whose value places a row along the x axis: `"_step"`, `"_runtime"`, `"_timestamp"` or a metric you logged yourself, such as `"epoch"`. A row without a number under it is left out.	`"_step"`
api	`wandb`.`Api`	The API that fetches a run named by its path. By default `wandb.Api()`.	`None`

Name	Type	Description
	dict of str to dict of str to MetricSeries	Metric by metric, and within a metric run by run, in the order the runs were given. A run that logged nothing under a metric is absent.

Name	Type	Description
	FileNotFoundError	If a run file named does not exist.
	ImportError	If a run is named by its server path and `wandb` is not installed.
	TypeError	If `runs` is none of the above.