Datasetpublic

openpecha/OCR-Handwritten_Tibetan_Cursive

Configuration: default Split: train Total Rows: 70,528 image_url Type: categorical Data Type: object Unique Values: 2 Value Distribution: Value Count Percentage Portrait 39,017 55.32% Landscape 31,511 44.68% Original README dataset_info: features: - name: image_name dtype: string - name: transcript dtype: string - name: image_url dtype: string - name: orientation dtype:… See the full description on the dataset page: https://huggingface.co/datasets/openpecha/OCR-Handwritten_Tibetan_Cursive.

sourceHugging Faceupdated 1y agoView on Hugging Face

0likes37downloads

Dataset Card

Configuration: `default`

Split: `train`

Total Rows: 70,528

`image_url`

—Type: categorical
—Data Type: object
—Unique Values: 2

Value Distribution:

Value	Count	Percentage
`Portrait`	39,017	55.32%
`Landscape`	31,511	44.68%

Original README

dataset_info: features:

—name: image_name dtype: string
—name: transcript dtype: string
—name: image_url dtype: string
—name: orientation dtype: string splits:
—name: train numbytes: 25328409 numexamples: 70528 downloadsize: 6047969 datasetsize: 25328409 configs:
—configname: default datafiles:
—split: train path: data/train-* ---