RIPS-Goog-23/FUNSD

---
dataset_info:
features:
- name: pixel_values
dtype:
array3_d:
shape:
- 3
- 224
- 224
dtype: float32
- name: input_ids
sequence: int64
- name: attention_mask
sequence: int64
- name: bbox
dtype:
array2_d:
shape:
- 512
- 4
dtype: int64
- name: labels
sequence: int64
- name: ner_tags
sequence:
class_label:
names:
'0': O
'1': B-HEADER
'2': I-HEADER
'3': B-QUESTION
'4': I-QUESTION
'5': B-ANSWER
'6': I-ANSWER
splits:
- name: test
num_bytes: 31847456
num_examples: 50
- name: train
num_bytes: 94872948
num_examples: 149
download_size: 5136310
dataset_size: 126720404
---
# Dataset Card for "FUNSD"

[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)

原始信息汇总

数据集概述

数据集特征

• pixel_values:

  • 数据类型: 3维数组
  • 形状: [3, 224, 224]
  • 数据类型: float32

• input_ids:

  • 数据类型: 序列, int64

• attention_mask:

  • 数据类型: 序列, int64

• bbox:

  • 数据类型: 2维数组
  • 形状: [512, 4]
  • 数据类型: int64

• labels:

  • 数据类型: 序列, int64

• ner_tags:

  • 数据类型: 序列
  • 类别标签:
    • 名称:
      • 0: O
      • 1: B-HEADER
      • 2: I-HEADER
      • 3: B-QUESTION
      • 4: I-QUESTION
      • 5: B-ANSWER
      • 6: I-ANSWER

数据集分割

• test:

  • 数据大小: 31847456 字节
  • 示例数量: 50

• train:

  • 数据大小: 94872948 字节
  • 示例数量: 149

数据集大小

• 下载大小: 5136310 字节
• 数据集总大小: 126720404 字节