(index, name)
| 133 | print_split_stats("test", 2) |
| 134 | |
| 135 | def build_dataset(index, name): |
| 136 | dataset = None |
| 137 | if splits[index + 1] > splits[index]: |
| 138 | documents = np.arange( |
| 139 | start=splits[index], stop=splits[index + 1], step=1, dtype=np.int32 |
| 140 | ) |
| 141 | dataset = PromptDataset( |
| 142 | name, |
| 143 | data_prefix, |
| 144 | documents, |
| 145 | input_ids_indexed_dataset, |
| 146 | attention_mask_indexed_dataset, |
| 147 | labels_indexed_dataset, |
| 148 | train_valid_test_num_samples[index], |
| 149 | seq_length, |
| 150 | seed, |
| 151 | ) |
| 152 | return dataset |
| 153 | |
| 154 | train_dataset = build_dataset(0, "train") |
| 155 | valid_dataset = build_dataset(1, "valid") |
no test coverage detected