metadata
license: bsd-3-clause
task_categories:
- text-generation
- feature-extraction
language:
- en
tags:
- theorem-proving
- formal-methods
- coq
- stdpp
- iris
size_categories:
- 1K<n<10K
Coq-Stdpp
Structured dataset of formalizations from the std++ library (used by Iris).
Dataset Description
- Source: iris/stdpp
- Entries: 4,982
- Files processed: 59
- License: BSD 3-Clause
Schema
| Column | Type | Description |
|---|---|---|
| fact | string | Full declaration (name, signature, body) |
| type | string | Lemma, Definition, Instance, Class, etc. |
| library | string | Sub-library (stdpp, stdpp_bitvector, stdpp_unstable) |
| imports | list | Import statements |
| filename | string | Source file path |
| symbolic_name | string | Declaration identifier |
| docstring | string | Documentation comment (9% coverage) |
Statistics
By Type
| Type | Count |
|---|---|
| Lemma | 3,374 |
| Instance | 1,000 |
| Definition | 311 |
| Class | 103 |
| Fixpoint | 93 |
| Inductive | 43 |
| Existing | 30 |
| Record | 18 |
| Other | 10 |
Example Entry
Use Cases
- Retrieval / RAG for Coq/Iris theorem proving
- Separation logic research
- Training embeddings for formal mathematics
Related Datasets
- phanerozoic/Coq-Iris - The Iris framework itself