uqlm.utils.dataloader.load_example_dataset#

uqlm.utils.dataloader.load_example_dataset(name, n=None, cols=None)#

Load a dataset for testing purposes.

Return type:

DataFrame

Parameters:
  • name (str) – The name of the dataset to load. Must be one of “svamp”, “gsm8k”, “ai2_arc”, “csqa”, “nq_open”, “popqa”

  • n (int, optional) – Number of rows to load from the dataset.

Returns:

The loaded dataset.

Return type:

pd.DataFrame

Example

>>> from uqlm.utils.dataloader import load_example_dataset
>>> df = load_example_dataset("gsm8k", n=1000)
>>> df.shape
(1000, 2)