Buddhist Sanskrit Canon Dataset

Overview

This dataset contains cleaned Sanskrit texts from the Buddhist Sanskrit Canon collected from the GRETIL (Göttingen Register of Electronic Texts in Indian Languages) digital library.

The texts were originally provided in various formats (plain text, HTML transformations, and TEI XML).
This dataset standardizes and cleans the texts for computational analysis and machine learning applications.

The dataset includes major Buddhist works such as:

Prajñāpāramitā texts
Vinaya literature
Avadāna collections
Mahāyāna sūtras
Dhāraṇī texts
Stotras and commentarial works
Yogācāra and Madhyamaka treatises

Examples of included texts:

Vimalakīrtinirdeśa Sūtra
Vasudhārā Dhāraṇī
Vinayaviniścaya (Upāli-paripṛcchā)
Viśvantarāvadāna
Vidyāsthānopama Sūtra
Ratnamālāvadāna
Divyāvadāna
Lalitavistara
Saddharmapuṇḍarīka Sūtra
Prajñāpāramitā literature

The texts are represented in IAST Roman Sanskrit, with optional Devanāgarī transliteration.

Source

The original texts were obtained from:

GRETIL – Göttingen Register of Electronic Texts in Indian Languages

https://gretil.sub.uni-goettingen.de/gretil.html

GRETIL provides electronic editions of classical Indian texts for scholarly use.

Dataset Maintainer

Maintained and curated by:

DataSpoof

Dataset Structure

Each text is processed into structured sections.

Example schema:

column	description
section_id	unique identifier for each text section
roman_sanskrit	cleaned Sanskrit text in IAST
devanagari	automatically generated Devanāgarī transliteration

Example:

section_id,roman_sanskrit,devanagari
vimal_1,buddhakṣetrapariśuddhinidānam namaḥ sarvātītapratyutpannānāgatebhyo...,बुद्धक्षेत्रपरिशुद्धिनिदानम् नमः सर्वातीत...
visv_3,so 'pareṇa samayena devyā sārdhaṃ krīḍati...,सोऽपरेण समयेन देव्याः सार्धं क्रीडति...