small_dense_structured_table

This dataset is generated syhthetically to create tables with following characteristics: 1. Empty cell percentage in following range [0,30] (Dense) 2. There is clear seperator between rows and columns (Structured). 3. 4 <= num rows <= 10, 2 <= num columns <= 6 (Small) ### Load the dataset ```python import io import pandas as pd from PIL import Image def bytes_to_image(self, image_bytes: bytes): return Image.open(io.BytesIO(image_bytes)) def parse_annotations(self, annotations: str) -> pd.DataFrame: return pd.read_json(StringIO(annotations), orient="records") test_data = load_dataset('nanonets/small_dense_structured_table', split='test') data_point = test_data[0] image, gt_table = ( bytes_to_image(data_point["images"]), parse_annotations(data_point["annotation"]), ) ```本数据集为合成生成，旨在构建具备如下特征的表格： 1. 空单元格占比处于[0, 30]区间内（密集型，Dense） 2. 行列间具备清晰分隔线（结构化，Structured） 3. 行数范围为4 ≤ 行数 ≤ 10，列数范围为2 ≤ 列数 ≤ 6（小型，Small） ### 数据集加载 python import io import pandas as pd from PIL import Image def bytes_to_image(self, image_bytes: bytes): return Image.open(io.BytesIO(image_bytes)) def parse_annotations(self, annotations: str) -> pd.DataFrame: return pd.read_json(StringIO(annotations), orient="records") test_data = load_dataset('nanonets/small_dense_structured_table', split='test') data_point = test_data[0] image, gt_table = ( bytes_to_image(data_point["images"]), parse_annotations(data_point["annotation"]), )

搜集汇总

背景与挑战

背景概述

该数据集由nanonets合成生成，旨在创建具有密集、结构化和小型特征的表格，其中空单元格比例范围为0-30%，行和列分隔清晰，且行数在4-10、列数在2-6之间。数据集采用Apache License 2.0许可，主要用于测试目的。