Datasets:

CCB
/

cis5300-language-models

configs:
  - config_name: cities
    data_files:
      - split: train
        path: cities/train-*
      - split: validation
        path: cities/validation-*
      - split: test
        path: cities/test-*

CIS 5300 Language Models Dataset

Dataset for Homework 3 of CIS 5300 (Natural Language Processing) at Penn.

Cities config

Country-of-origin classification over short city-name strings, drawn from nine countries (Afghanistan, China, Germany, Finland, France, India, Iran, Pakistan, South Africa).

from datasets import load_dataset
cities = load_dataset("CCB/cis5300-language-models", "cities")

Split	Rows	Has labels?
`train`	12,392	yes
`validation`	1,548	yes
`test`	1,531	no (students predict)

Each row has city, country (ISO code), and country_name. For the test split, country and country_name are empty strings -- ground truth is held back so students must submit predictions for grading.

Raw corpus files

For character-level language modeling and cross-domain perplexity work:

File	Size	Description
`shakespeare_input.txt`	4.5 MB	Complete Shakespeare works
`shakespeare_sonnets.txt`	9 KB	Held-out Shakespeare sonnets
`modern_english_sample.txt`	4.4 KB	Modern-English prose for cross-domain perplexity

Fetch with huggingface_hub.hf_hub_download(..., repo_type="dataset").

Licensing and attribution

shakespeare_input.txt and shakespeare_sonnets.txt are in the public domain.
modern_english_sample.txt is an excerpt from the English Wikipedia article "Natural language processing" (https://en.wikipedia.org/wiki/Natural_language_processing), used under the Creative Commons Attribution-ShareAlike 4.0 license (https://creativecommons.org/licenses/by-sa/4.0/). The excerpt covers the article's introduction and history sections, with Wikipedia section headers removed so the text reads as continuous prose. See ATTRIBUTIONS.md for details.
The cities config is released under CC BY 4.0.