(self)
| 27 | """Each of these test datasets has 100 examples""" |
| 28 | |
| 29 | def test_loading_data_args(self): |
| 30 | dataset_mixer = { |
| 31 | "HuggingFaceH4/testing_alpaca_small": 0.5, |
| 32 | "HuggingFaceH4/testing_self_instruct_small": 0.3, |
| 33 | "HuggingFaceH4/testing_codealpaca_small": 0.2, |
| 34 | } |
| 35 | data_args = DataArguments(dataset_mixer=dataset_mixer) |
| 36 | datasets = get_datasets(data_args, columns_to_keep=["prompt", "completion"]) |
| 37 | self.assertEqual(len(datasets["train"]), 100) |
| 38 | self.assertEqual(len(datasets["test"]), 300) |
| 39 | |
| 40 | def test_loading_data_dict(self): |
| 41 | dataset_mixer = { |
nothing calls this directly
no test coverage detected