Evaluate

You are viewing v0.4.4 version. A newer version v0.4.6 is available.

Join the Hugging Face community

and get access to the augmented documentation experience

Collaborate on models, datasets and Spaces

Faster examples with accelerated inference

Switch between documentation themes

to get started

Saving methods

Methods for saving evaluations results:

Save

evaluate.save

< source >

( path_or_file **data )

Parameters

path_or_file (str) — Path or file to store the file. If only a folder is provided the results file will be saved in the format "result-%Y_%m_%d-%H_%M_%S.json".

Saves results to a JSON file. Also saves system information such as current time, current commit hash if inside a repository, and Python system information.

Example:

>>> import evaluate
>>> result = {"bleu": 0.7}
>>> params = {"model": "gpt-2"}
>>> evaluate.save("./results/", **result, **params)

< > Update on GitHub

←Loading methods Hub methods→