fbaigt/schema-to-json

---
license: gpl-3.0
configs:
- config_name: chemtables
data_files:
- split: train
path: chemtables/train-*
- split: validation
path: chemtables/validation-*
- split: test
path: chemtables/test-*
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*
- config_name: discomat
data_files:
- split: train
path: discomat/train-*
- split: validation
path: discomat/validation-*
- split: test
path: discomat/test-*
- config_name: mltables
data_files:
- split: train
path: mltables/train-*
- split: validation
path: mltables/validation-*
- split: test
path: mltables/test-*
dataset_info:
- config_name: chemtables
features:
- name: paper_id
dtype: string
- name: table_id
dtype: string
- name: table_code
dtype: string
- name: sup_text
dtype: string
- name: target_cells
sequence:
- name: cell_value
dtype: string
- name: cell_raw
dtype: string
- name: cell_index
dtype: string
- name: cell_row_idx
dtype: int32
- name: cell_col_idx
dtype: int32
- name: gold_json_records
sequence:
- name: cell_index
dtype: string
- name: cell_record
dtype: string
splits:
- name: train
num_bytes: 92180
num_examples: 9
- name: validation
num_bytes: 39374
num_examples: 3
- name: test
num_bytes: 117148
num_examples: 14
download_size: 124818
dataset_size: 248702
- config_name: default
features:
- name: paper_id
dtype: string
- name: table_id
dtype: string
- name: table_code
dtype: string
- name: sup_text
dtype: string
- name: target_cells
sequence:
- name: cell_value
dtype: string
- name: cell_raw
dtype: string
- name: cell_index
dtype: string
- name: cell_row_idx
dtype: int32
- name: cell_col_idx
dtype: int32
- name: gold_json_records
sequence:
- name: cell_index
dtype: string
- name: cell_record
dtype: string
splits:
- name: train
num_bytes: 78484
num_examples: 9
- name: validation
num_bytes: 37457
num_examples: 3
- name: test
num_bytes: 113119
num_examples: 14
download_size: 122465
dataset_size: 229060
- config_name: discomat
features:
- name: paper_id
dtype: string
- name: table_id
dtype: string
- name: table_code
dtype: string
- name: sup_text
dtype: string
- name: target_cells
sequence:
- name: cell_value_processed
dtype: string
- name: i
dtype: int32
- name: j
dtype: int32
- name: k
dtype: int32
- name: gold_json_records
sequence:
- name: cell_index
sequence: int32
length: 3
- name: cell_record
dtype: string
splits:
- name: train
num_bytes: 2300237
num_examples: 500
- name: validation
num_bytes: 2300237
num_examples: 500
- name: test
num_bytes: 2366158
num_examples: 487
download_size: 1430344
dataset_size: 6966632
- config_name: mltables
features:
- name: paper_id
dtype: string
- name: table_id
dtype: string
- name: table_code
dtype: string
- name: sup_text
dtype: string
- name: target_cells
sequence:
- name: cell_value
dtype: string
- name: cell_raw
dtype: string
- name: cell_value_char_idx_start
dtype: int32
- name: cell_value_char_idx_end
dtype: int32
- name: cell_raw_char_idx_start
dtype: int32
- name: cell_raw_char_idx_end
dtype: int32
- name: gold_json_records
sequence:
- name: cell_char_index
sequence: int32
length: 2
- name: cell_record
dtype: string
splits:
- name: train
num_bytes: 696651
num_examples: 43
- name: validation
num_bytes: 150816
num_examples: 11
- name: test
num_bytes: 1248693
num_examples: 68
download_size: 605737
dataset_size: 2096160
---

原始信息汇总

数据集概述

许可证

• GPL-3.0

配置信息

chemtables

• 数据文件路径
  • 训练集: chemtables/train-*
  • 验证集: chemtables/validation-*
  • 测试集: chemtables/test-*
• 特征
  • paper_id: 字符串
  • table_id: 字符串
  • table_code: 字符串
  • sup_text: 字符串
  • target_cells: 序列
    • cell_value: 字符串
    • cell_raw: 字符串
    • cell_index: 字符串
    • cell_row_idx: 整数32位
    • cell_col_idx: 整数32位
  • gold_json_records: 序列
    • cell_index: 字符串
    • cell_record: 字符串
• 分割信息
  • 训练集: 92180字节, 9个样本
  • 验证集: 39374字节, 3个样本
  • 测试集: 117148字节, 14个样本
• 下载大小: 124818字节
• 数据集大小: 248702字节

default

• 数据文件路径
  • 训练集: data/train-*
  • 验证集: data/validation-*
  • 测试集: data/test-*
• 特征
  • paper_id: 字符串
  • table_id: 字符串
  • table_code: 字符串
  • sup_text: 字符串
  • target_cells: 序列
    • cell_value: 字符串
    • cell_raw: 字符串
    • cell_index: 字符串
    • cell_row_idx: 整数32位
    • cell_col_idx: 整数32位
  • gold_json_records: 序列
    • cell_index: 字符串
    • cell_record: 字符串
• 分割信息
  • 训练集: 78484字节, 9个样本
  • 验证集: 37457字节, 3个样本
  • 测试集: 113119字节, 14个样本
• 下载大小: 122465字节
• 数据集大小: 229060字节

discomat

• 数据文件路径
  • 训练集: discomat/train-*
  • 验证集: discomat/validation-*
  • 测试集: discomat/test-*
• 特征
  • paper_id: 字符串
  • table_id: 字符串
  • table_code: 字符串
  • sup_text: 字符串
  • target_cells: 序列
    • cell_value_processed: 字符串
    • i: 整数32位
    • j: 整数32位
    • k: 整数32位
  • gold_json_records: 序列
    • cell_index: 序列, 整数32位, 长度3
    • cell_record: 字符串
• 分割信息
  • 训练集: 2300237字节, 500个样本
  • 验证集: 2300237字节, 500个样本
  • 测试集: 2366158字节, 487个样本
• 下载大小: 1430344字节
• 数据集大小: 6966632字节

mltables

• 数据文件路径
  • 训练集: mltables/train-*
  • 验证集: mltables/validation-*
  • 测试集: mltables/test-*
• 特征
  • paper_id: 字符串
  • table_id: 字符串
  • table_code: 字符串
  • sup_text: 字符串
  • target_cells: 序列
    • cell_value: 字符串
    • cell_raw: 字符串
    • cell_value_char_idx_start: 整数32位
    • cell_value_char_idx_end: 整数32位
    • cell_raw_char_idx_start: 整数32位
    • cell_raw_char_idx_end: 整数32位
  • gold_json_records: 序列
    • cell_char_index: 序列, 整数32位, 长度2
    • cell_record: 字符串
• 分割信息
  • 训练集: 696651字节, 43个样本
  • 验证集: 150816字节, 11个样本
  • 测试集: 1248693字节, 68个样本
• 下载大小: 605737字节
• 数据集大小: 2096160字节