FudanSELab/codegen4libs
repo for the paper titled “CodeGen4Libs: A Two-Stage Approach for Library-Oriented Code Generation”
This project provides a dataset and models for generating Python code, specifically focusing on how to use libraries. It takes natural language descriptions or existing code snippets as input and outputs Python code, including relevant library imports. This is for software developers or researchers working on code generation and understanding large language models for programming tasks.
No commits in the last 6 months.
Use this if you are a developer training or evaluating models for code generation, especially those that need to correctly handle library usage and imports in Python.
Not ideal if you are a non-developer seeking an end-user application to generate code from natural language without needing to interact with models or datasets directly.
Stars
14
Forks
1
Language
Python
License
—
Category
Last pushed
Oct 04, 2023
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ai-coding/FudanSELab/codegen4libs"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
howerj/dbcc
CAN DBC to C (and CSV, JSON and XML) compiler using the mpc parser combinator library
JhnW/devana
Python package to parse and generate C/C++ code as context aware preprocessor.
biojppm/regen
Easy C++ reflection and code generation
SoftSec-KAIST/CodeAlchemist
CodeAlchemist: Semantics-Aware Code Generation to Find Vulnerabilities in JavaScript Engines (NDSS '19)
Samsung/UTopia
UT based automated fuzz driver generation