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

torch format for dataset

Browse Source
This commit is contained in:
Michał Kozłowski 2023-01-24 18:18:50 +01:00
parent 5fca831ccd
commit 0466f8f60e
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
utils/donut_dataset.py
View File

@ -48,7 +48,7 @@ class DonutDataset(Dataset):
self.added_tokens = added_tokens self.added_tokens = added_tokens
self.dataset = dataset self.dataset = dataset
self.dataset_length = len(self.dataset) self.dataset_length = len(self.dataset.with_format("torch"))
self.gt_token_sequences = [] self.gt_token_sequences = []
for sample in self.dataset: for sample in self.dataset: