afnio.utils.datasets.facility_support#

Meta’s Facility Support Analyzer dataset.

Classes

FacilitySupport(split[, root])

The Meta Facility Support Analyzer dataset consists of 200 real-world emails or messages sent in enterprise settings related to facility maintenance or support requests.
class afnio.utils.datasets.facility_support.FacilitySupport(split, root=None)[source]#

Bases: Dataset

The Meta Facility Support Analyzer dataset consists of 200 real-world emails or messages sent in enterprise settings related to facility maintenance or support requests. Each example is annotated with:

  • urgency (low, medium, high)

  • sentiment (negative, neutral, positive)

  • relevant service request categories (e.g., cleaning, IT support, maintenance)

The dataset is split into train, validation, and test sets with a 33%/33%/34% ratio. The split is deterministic, ensuring reproducibility across different runs.

Parameters:

  • split (str) – The dataset split to load. Must be either “train”, “val”, or “test”.

  • root (Union[str, Path], optional) – The root directory where JSON files are stored. Defaults to None.

download()[source]#

Download the Facility Support data if it doesn’t exist already.

extra_repr()[source]#

Returns additional information about the dataset.

mirrors = ['https://raw.githubusercontent.com/meta-llama/llama-prompt-ops/refs/heads/main/use-cases/facility-support-analyzer/']#
property raw_folder: str#
resources = [('dataset.json', '530dc66b1b07c9b15b19f08891e9bfa0')]#