mirror of
https://github.com/blackboxprogramming/lucidia.git
synced 2026-03-17 03:57:13 -05:00
Add data/build_pretrain.py with pretrain dataset builder stub
This commit is contained in:
committed by
GitHub
parent
91fb34569e
commit
0576512be0
6
lucidia_llm/data/build_pretrain.py
Normal file
6
lucidia_llm/data/build_pretrain.py
Normal file
@@ -0,0 +1,6 @@
|
|||||||
|
"""Data builder for pretraining dataset."""
|
||||||
|
|
||||||
|
|
||||||
|
def build_pretrain_dataset(input_paths, output_path):
|
||||||
|
"""Construct the pretraining dataset from raw inputs (placeholder)."""
|
||||||
|
raise NotImplementedError("Pretrain dataset builder not implemented")
|
||||||
Reference in New Issue
Block a user