README.md

---
dataset_info:
  features:
  - name: chunk_id
    dtype: string
  - name: text
    dtype: string
  - name: metadata
    struct:
    - name: Header 1
      dtype: string
    - name: Header 2
      dtype: string
    - name: image_references
      sequence: string
    - name: images_base64
      sequence: string
    - name: start_index
      dtype: int64
  - name: source_filename
    dtype: string
  splits:
  - name: train
    num_bytes: 19469481
    num_examples: 271
  download_size: 15393879
  dataset_size: 19469481
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
language:
- en
tags:
- africa
- history
- vllm
- RAG
size_categories:
- 1K<n<10K
---