summaryrefslogtreecommitdiff
path: root/text_recognizer/data/iam_extended_paragraphs.py
diff options
context:
space:
mode:
authorGustaf Rydholm <gustaf.rydholm@gmail.com>2021-04-22 08:15:58 +0200
committerGustaf Rydholm <gustaf.rydholm@gmail.com>2021-04-22 08:15:58 +0200
commit1ca8b0b9e0613c1e02f6a5d8b49e20c4d6916412 (patch)
tree5e610ac459c9b254f8826e92372346f01f8e2412 /text_recognizer/data/iam_extended_paragraphs.py
parentffa4be4bf4e3758e01d52a9c1f354a05a90b93de (diff)
Fixed training script, able to train vqvae
Diffstat (limited to 'text_recognizer/data/iam_extended_paragraphs.py')
-rw-r--r--text_recognizer/data/iam_extended_paragraphs.py15
1 files changed, 12 insertions, 3 deletions
diff --git a/text_recognizer/data/iam_extended_paragraphs.py b/text_recognizer/data/iam_extended_paragraphs.py
index d2529b4..2380660 100644
--- a/text_recognizer/data/iam_extended_paragraphs.py
+++ b/text_recognizer/data/iam_extended_paragraphs.py
@@ -10,18 +10,27 @@ from text_recognizer.data.iam_synthetic_paragraphs import IAMSyntheticParagraphs
class IAMExtendedParagraphs(BaseDataModule):
def __init__(
self,
- batch_size: int = 128,
+ batch_size: int = 16,
num_workers: int = 0,
train_fraction: float = 0.8,
augment: bool = True,
+ word_pieces: bool = False,
) -> None:
super().__init__(batch_size, num_workers)
self.iam_paragraphs = IAMParagraphs(
- batch_size, num_workers, train_fraction, augment,
+ batch_size,
+ num_workers,
+ train_fraction,
+ augment,
+ word_pieces,
)
self.iam_synthetic_paragraphs = IAMSyntheticParagraphs(
- batch_size, num_workers, train_fraction, augment,
+ batch_size,
+ num_workers,
+ train_fraction,
+ augment,
+ word_pieces,
)
self.dims = self.iam_paragraphs.dims