nemo_automodel.components.datasets.llm.mock_seq_cls#
Module Contents#
Classes#
Mock dataset for sequence classification functional tests. |
API#
- class nemo_automodel.components.datasets.llm.mock_seq_cls.MockSequenceClassificationDataset(
- *,
- num_samples: int = 64,
- num_labels: int = 2,
- vocab_size: int = 256,
- max_seq_len: int = 32,
- seed: int = 0,
- tokenizer=None,
Bases:
torch.utils.data.DatasetMock dataset for sequence classification functional tests.
Generates random token sequences with binary labels. Does not require a tokenizer or network access.
Initialization
- __len__()#
- __getitem__(idx)#