Datasets:

nola-ai
/

capitals-demo

proper_noun stringlengths 2 57	first_letter stringclasses 26 values
A Coruna	A
A Estrada	A
Aabenraa	A
Aachen	A
Aalborg	A
Aalen	A
Aalsmeer	A
Aalst	A
Aalten	A
Aalter	A
Aarau	A
Aarschot	A
Aba	A
Abadan	A
Abadeh	A
Abadia De Goias	A
Abaete	A
Abaetetuba	A
Abakaliki	A
Abakan	A
Abala	A
Abalak	A
Abancay	A
Abano Terme	A
Abare	A
Abashiri	A
Abasolo	A
Abay	A
Abaza	A
Abbeville	A
Abbiategrasso	A
Abbotsford	A
Abbottabad	A
Abdanan	A
Abderafi	A
Abdu Rahiman Nagar	A
Abdulino	A
Abeche	A
Abejorral	A
Abelardo Luz	A
Abengourou	A
Abeokuta	A
Abepura	A
Aberdare	A
Aberdeen	A
Abergele	A
Aberystwyth	A
Abha	A
Abhar	A
Abhayapuri	A
Abidjan	A
Abiko	A
Abilene	A
Abim	A
Abingdon	A
Abington	A
Abinsk	A
Abiy Adi	A
Abnub	A
Abobo	A
Abohar	A
Aboisso	A
Aboisso Comoe	A
Aboka	A
Abomey	A
Abomey-calavi	A
Abomsa	A
Abong Mbang	A
Abony	A
Abou El Hassan	A
Abovyan	A
Abqaiq	A
Abrama	A
Abrantes	A
Abreu E Lima	A
Abreus	A
Abrisham	A
Abu	A
Abu 'arish	A
Abu Al Matamir	A
Abu An Numrus	A
Abu Dhabi	A
Abu Ghurayb	A
Abu Hammad	A
Abu Hayl	A
Abu Hummus	A
Abu Jibeha	A
Abu Kabir	A
Abu Qurqas	A
Abu Road	A
Abu Suweir-el-mahatta	A
Abu Tij	A
Abu Tisht	A
Abu Zabad	A
Abucay	A
Abuja	A
Abuko	A
Aburi	A
Abuyog	A
Abyei	A

End of preview. Expand in Data Studio

Summary

capitals-demo is a small educational dataset created for teaching and demonstration purposes. It was built using the open-source geonamescache Python library and contains approximately 30,200 examples.
Each example pairs a geographic location name with its first capital letter. It was created for our ATōMIZER tutorial on Fine-tuning small language models for new tasks.

Supported Tasks:

Text Classification
Token Extraction / Character-Level Prediction

Languages: English
Version: 1.0

Owner: NOLA-AI

Dataset Overview

capitals-demo demonstrates how to create a simple supervised learning dataset from an existing Python resource.
The data consists of geographic proper nouns mapped to their initial letters, intended for tutorials and educational use in text processing or beginner machine learning tasks.

Dataset

Structure

Each record includes the following fields:

Field	Description
`proper_noun`	A geographic location name
`first_letter`	The uppercase first letter of the location name

Example

{
  "proper_noun": "Barcelona",
  "first_letter": "B"
}

Downloads last month: 5

Size of downloaded dataset files:

395 kB

Size of the auto-converted Parquet files:

310 kB

Number of rows:

30,238