metadata
dataset_info:
features:
- name: guide
struct:
- name: title
dtype: string
- name: chapters
list:
- name: chapter
dtype: int64
- name: title
dtype: string
- name: sections
list:
- name: number
dtype: int64
- name: subsections
list:
- name: number
dtype: int64
- name: topic
dtype: string
- name: content
dtype: string
- name: subsection
dtype: int64
- name: example
dtype: string
- name: tip
dtype: string
- name: note
dtype: string
- name: exercise
dtype: string
- name: practice
dtype: string
- name: technique
dtype: string
- name: strategy
dtype: string
- name: best_practice
dtype: string
- name: approach
dtype: string
- name: rule_of_thumb
dtype: string
- name: guideline
dtype: string
- name: process
dtype: string
- name: advice
dtype: string
- name: guidelines
dtype: string
- name: style
dtype: string
- name: tips
dtype: string
- name: caution
dtype: string
- name: benefit
dtype: string
- name: insight
dtype: string
- name: best_practices
dtype: string
- name: principle
dtype: string
- name: methodology
dtype: string
- name: procedure
dtype: string
- name: styleguide
dtype: string
- name: editing
dtype: string
- name: perspective
dtype: string
- name: rhetoric
dtype: string
- name: section
dtype: int64
splits:
- name: train
num_bytes: 33377
num_examples: 1
download_size: 65012
dataset_size: 33377
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
license: mit
language:
- en
source_datasets: TuringsSolutions/NYTWritingStyleGuide
pszemraj/NYTWritingStyleGuide-parsed
Instead of the alien and unparsable/unloadable format of the original, this can be loaded as a dataset.