`dawsonia.ml.data.reader`#

Dataset reader and process.

Module Contents#

Classes#

Dataset

Dataset class to read images and sentences from base (raw files)

Data#

`logger`
`PartitionName`

API#

dawsonia.ml.data.reader.logger#: ‘getLogger(…)’

dawsonia.ml.data.reader.PartitionName: typing_extensions.TypeAlias#: None

class dawsonia.ml.data.reader.Dataset(source, name)#

Dataset class to read images and sentences from base (raw files)

Initialization

create_unified_dataset(target: pathlib.Path, image_input_size, max_text_length)#

read_partitions()#: Read images and sentences from dataset.

save_partitions(target, image_input_size, max_text_length, debug=False, partitions=None)#: Save images and sentences from dataset.

_init_dataset()#

_shuffle(*ls)#

_hdsr14_car_a()#: ICFHR 2014 Competition on Handwritten Digit String Recognition in Challenging Datasets dataset reader.

_hdsr14_car_b()#: ICFHR 2014 Competition on Handwritten Digit String Recognition in Challenging Datasets dataset reader.

_read_orand_partitions(basedir, type_f)#: ICFHR 2014 Competition on Handwritten Digit String Recognition in Challenging Datasets dataset reader.

_hdsr14_cvl()#: ICFHR 2014 Competition on Handwritten Digit String Recognition in Challenging Datasets dataset reader.

_dida()#

DIDA dataset 12k digit string images and 12k digit string labels

Reference

Huseyin Kusetogullari, Amir Yavariabdi, Johan Hall, Niklas Lavesson, DIDA: The largest historical handwritten digit dataset with 250k digits, June 2021. Accessed on: June 13, 2021. Available: didadataset/DIDA.

_bentham()#: Bentham dataset reader.

_iam()#: IAM dataset reader.

_rimes()#: Rimes dataset reader.

_saintgall()#: Saint Gall dataset reader.

_washington()#: Washington dataset reader.

static check_text(data, max_text_length=128)#: Checks if the text has more characters instead of punctuation marks.

dawsonia.ml.data.reader#