From 644c41901fd12570fd0a50a24d09476913b46a6b Mon Sep 17 00:00:00 2001 From: Robert Knight Date: Sat, 10 Feb 2024 08:38:11 +0000 Subject: [PATCH] Add a note about which tasks each dataset is for --- ocrs_models/datasets.py | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/ocrs_models/datasets.py b/ocrs_models/datasets.py index 961ac0d..1d90bb9 100644 --- a/ocrs_models/datasets.py +++ b/ocrs_models/datasets.py @@ -141,7 +141,7 @@ class SizedDataset(Dataset, Sized): class DDI100(SizedDataset): """ - Distorted Document Images (DDI-100) dataset. + Distorted Document Images (DDI-100) dataset for text detection. See https://github.com/machine-intelligence-laboratory/DDI-100 and https://arxiv.org/abs/1912.11658. @@ -230,7 +230,7 @@ def reorder_quad(quad): class HierText(SizedDataset): """ - HierText dataset. + HierText dataset for text detection. See https://github.com/google-research-datasets/hiertext and https://arxiv.org/abs/2203.15143.