Megatron-DeepSpeed
0c820648 - update merge_preprocessed_data to use distributed merge (#82)

Commit
4 years ago
update merge_preprocessed_data to use distributed merge (#82) * update merge_preprocessed_data to use parallel merge * indexed_dataset: add docstrings to merge and gather methods * merge_preprocessed_data: tweak interface, add documentation * merge: improvements after testing * tests: serial and distributed merge * avoid setting pythonpath within script * merge script: fix typo in usage comments * print default backend when not set in distributed merge
Author
Parents
Loading