s444415/donut
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

80% split

Browse Source
This commit is contained in:
Michał Kozłowski 2023-01-25 22:18:39 +01:00
parent a436a38ddc
commit dda04a1d5c
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
train_stream.py
View File

@ -34,7 +34,7 @@ def main(config, hug_token):
added_tokens = []
dataset = load_dataset(config.dataset_path)
dataset = load_dataset(config.dataset_path, split="train[:80%]")
dataset = dataset.train_test_split(test_size=0.1)
train_dataset = DonutDataset(