Dataset Card for Dataset Name

This dataset card aims to be a base template for new datasets. It has been generated using this raw template.

  Dataset Details





  Dataset Description

Curated by: [More Information Needed] Funded by [optional]: [More Information Needed] Shared by [optional]: [More Information Needed] Language(s) (NLP): [More Information Needed] License: [More Information Needed]

  Dataset Sources [optional]

Repository: [More… See the full description on the dataset page: https://huggingface.co/datasets/systemk/codenet.

Project CodeNet is a large-scale dataset with approximately 14 million code samples, each of which is an intended solution to one of 4000 coding problems. The code samples are written in over 50 programming languages (although the dominant languages are C++, C, Python, and Java) and they are annotated with a rich set of information, such as its code size, memory footprint, cpu run time, and status, which indicates acceptance or error types. The dataset is accompanied by a repository, where we provide a set of tools to aggregate codes samples based on user criteria and to transform code samples into token sequences, simplified parse trees and other code graphs. A detailed discussion of Project CodeNet is available in this paper.

The rich annotation of Project CodeNet enables research in code search, code completion, code-code translation, and a myriad of other use cases. We also extracted several benchmarks in Python, Java and C++ to drive innovation in deep learning and machine learning models in code classification and code similarity.

Citation @inproceedings{puri2021codenet, author = {Ruchir Puri and David Kung and Geert Janssen and Wei Zhang and Giacomo Domeniconi and Vladmir Zolotov and Julian Dolby and Jie Chen and Mihir Choudhury and Lindsey Decker and Veronika Thost and Luca Buratti and Saurabh Pujar and Ulrich Finkler}, title = {Project CodeNet: A Large-Scale AI for Code Dataset for Learning a Diversity of Coding Tasks}, year = {2021}, }

iidai/codenet dataset hosted on Hugging Face and contributed by the HF Datasets community

iNeil77/CodeNet dataset hosted on Hugging Face and contributed by the HF Datasets community

📊 Dataset Card for 🏆 CodeNet-16K

  Dataset Summary

The 🏆 CodeNet-16K dataset consists of 16,500 Python attempts from the CodeNet dataset, which have been carefully filtered and deduplicated to create a high-quality dataset for code generation tasks. The dataset includes problem descriptions, input/output descriptions, and sample test cases for each problem.

  Dataset Details





  Dataset Sources

Repository:… See the full description on the dataset page: https://huggingface.co/datasets/sumuks/CodeNet-16K.

Artifact repository for the paper Lost in Translation: A Study of Bugs Introduced by Large Language Models while Translating Code, accepted at ICSE 2024, Lisbon, Portugal. Authors are Rangeet Pan* Ali Reza Ibrahimzada*, Rahul Krishna, Divya Sankar, Lambert Pougeum Wassi, Michele Merler, Boris Sobolev, Raju Pavuluri, Saurabh Sinha, and Reyhaneh Jabbarvand.

Install

This repository contains the source code for reproducing the results in our paper. Please start by cloning this repository:

git clone https://github.com/Intelligent-CAT-Lab/PLTranslationEmpirical

We recommend using a virtual environment for running the scripts. Please download conda 23.11.0 from this link. You can create a virtual environment using the following command:

conda create -n plempirical python=3.10.13

After creating the virtual environment, you can activate it using the following command:

conda activate plempirical

You can run the following command to make sure that you are using the correct version of Python:

python3 --version && pip3 --version

Dependencies

To install all software dependencies, please execute the following command:

pip3 install -r requirements.txt

As for hardware dependencies, we used 16 NVIDIA A100 GPUs with 80GBs of memory for inferencing models. The models can be inferenced on any combination of GPUs as long as the reader can properly distribute the model weights across the GPUs. We did not perform weight distribution since we had enough memory (80 GB) per GPU.

Moreover, for compiling and testing the generated translations, we used Python 3.10, g++ 11, GCC Clang 14.0, Java 11, Go 1.20, Rust 1.73, and .Net 7.0.14 for Python, C++, C, Java, Go, Rust, and C#, respectively. Overall, we recommend using a machine with Linux OS and at least 32GB of RAM for running the scripts.

For running scripts of alternative approaches, you need to make sure you have installed C2Rust, CxGO, and Java2C# on your machine. Please refer to their repositories for installation instructions. For Java2C#, you need to create a .csproj file like below:

Dataset

We uploaded the dataset we used in our empirical study to Zenodo. The dataset is organized as follows:

1. CodeNet
2. AVATAR
3. Evalplus
4. Apache Commons-CLI
5. Click

Please download and unzip the dataset.zip file from Zenodo. After unzipping, you should see the following directory structure:

PLTranslationEmpirical
├── dataset
  ├── codenet
  ├── avatar
  ├── evalplus
  ├── real-life-cli
├── ...

The structure of each dataset is as follows:

1. CodeNet & Avatar: Each directory in these datasets correspond to a source language where each include two directories Code and TestCases for code snippets and test cases, respectively. Each code snippet has an id in the filename, where the id is used as a prefix for test I/O files.

2. Evalplus: The source language code snippets follow a similar structure as CodeNet and Avatar. However, as a one time effort, we manually created the test cases in the target Java language inside a maven project, evalplus_java. To evaluate the translations from an LLM, we recommend moving the generated Java code snippets to the src/main/java directory of the maven project and then running the command mvn clean test surefire-report:report -Dmaven.test.failure.ignore=true to compile, test, and generate reports for the translations.

3. Real-life Projects: The real-life-cli directory represents two real-life CLI projects from Java and Python. These datasets only contain code snippets as files and no test cases. As mentioned in the paper, the authors manually evaluated the translations for these datasets.

Scripts

We provide bash scripts for reproducing our results in this work. First, we discuss the translation script. For doing translation with a model and dataset, first you need to create a .env file in the repository and add the following:

OPENAI_API_KEY=

1. Translation with GPT-4: You can run the following command to translate all Python -> Java code snippets in codenet dataset with the GPT-4 while top-k sampling is k=50, top-p sampling is p=0.95, and temperature=0.7:

bash scripts/translate.sh GPT-4 codenet Python Java 50 0.95 0.7 0

2. Translation with CodeGeeX: Prior to running the script, you need to clone the CodeGeeX repository from here and use the instructions from their artifacts to download their model weights. After cloning it inside PLTranslationEmpirical and downloading the model weights, your directory structure should be like the following:

PLTranslationEmpirical
├── dataset
  ├── codenet
  ├── avatar
  ├── evalplus
  ├── real-life-cli
├── CodeGeeX
  ├── codegeex
  ├── codegeex_13b.pt # this file is the model weight
  ├── ...
├── ...

You can run the following command to translate all Python -> Java code snippets in codenet dataset with the CodeGeeX while top-k sampling is k=50, top-p sampling is p=0.95, and temperature=0.2 on GPU gpu_id=0:

bash scripts/translate.sh CodeGeeX codenet Python Java 50 0.95 0.2 0

3. For all other models (StarCoder, CodeGen, LLaMa, TB-Airoboros, TB-Vicuna), you can execute the following command to translate all Python -> Java code snippets in codenet dataset with the StarCoder|CodeGen|LLaMa|TB-Airoboros|TB-Vicuna while top-k sampling is k=50, top-p sampling is p=0.95, and temperature=0.2 on GPU gpu_id=0:

bash scripts/translate.sh StarCoder codenet Python Java 50 0.95 0.2 0

4. For translating and testing pairs with traditional techniques (i.e., C2Rust, CxGO, Java2C#), you can run the following commands:

bash scripts/translate_transpiler.sh codenet C Rust c2rust fix_report
bash scripts/translate_transpiler.sh codenet C Go cxgo fix_reports
bash scripts/translate_transpiler.sh codenet Java C# java2c# fix_reports
bash scripts/translate_transpiler.sh avatar Java C# java2c# fix_reports

5. For compile and testing of CodeNet, AVATAR, and Evalplus (Python to Java) translations from GPT-4, and generating fix reports, you can run the following commands:

bash scripts/test_avatar.sh Python Java GPT-4 fix_reports 1
bash scripts/test_codenet.sh Python Java GPT-4 fix_reports 1
bash scripts/test_evalplus.sh Python Java GPT-4 fix_reports 1

6. For repairing unsuccessful translations of Java -> Python in CodeNet dataset with GPT-4, you can run the following commands:

bash scripts/repair.sh GPT-4 codenet Python Java 50 0.95 0.7 0 1 compile
bash scripts/repair.sh GPT-4 codenet Python Java 50 0.95 0.7 0 1 runtime
bash scripts/repair.sh GPT-4 codenet Python Java 50 0.95 0.7 0 1 incorrect

7. For cleaning translations of open-source LLMs (i.e., StarCoder) in codenet, you can run the following command:

bash scripts/clean_generations.sh StarCoder codenet

Please note that for the above commands, you can change the dataset and model name to execute the same thing for other datasets and models. Moreover, you can refer to /prompts for different vanilla and repair prompts used in our study.

Artifacts

Please download the artifacts.zip file from our Zenodo repository. We have organized the artifacts as follows:

1. RQ1 - Translations: This directory contains the translations

sumuks/CodeNet-24K dataset hosted on Hugging Face and contributed by the HF Datasets community

CoQuIR/CodeNet-B dataset hosted on Hugging Face and contributed by the HF Datasets community

Explore the historical Whois records related to hs-code.net (Domain). Get insights into ownership history and changes over time.

Ovaj skup podataka uključuje finansijske izvještaje, račune i blokade, te nekretnine. Podaci uključuju prihode, rashode, dobit, imovinu, obaveze i informacije o nekretninama u vlasništvu kompanije. Finansijski podaci, finansijski sažetak, sažetak kompanije, preduzetnik, zanatlija, udruženje, poslovni subjekti.

This dataset is about countries in Caribbean. It has 13 rows. It features 4 columns: ISO 3 country code, self-employed workers, and net migration.

Explore the historical Whois records related to chess-code.net (Domain). Get insights into ownership history and changes over time.

Dataset Card for CodeContests

  Dataset Summary

CodeContests is a competitive programming dataset for machine-learning. This dataset was used when training AlphaCode. It consists of programming problems, from a variety of sources:

Site URL Source

Aizu https://judge.u-aizu.ac.jp CodeNet

AtCoder https://atcoder.jp CodeNet

CodeChef https://www.codechef.com description2code

Codeforces https://codeforces.com description2code and Codeforces

HackerEarth… See the full description on the dataset page: https://huggingface.co/datasets/deepmind/code_contests.

Explore the historical Whois records related to c-code.net (Domain). Get insights into ownership history and changes over time.

Explore the historical Whois records related to short-code.net (Domain). Get insights into ownership history and changes over time.

Explore the historical Whois records related to free-code.net (Domain). Get insights into ownership history and changes over time.

This is dataset is extracted from CodeNet, python only. I merged the data into one single table, including metadata, problem description, test input output.

small: accepted status only big: all status, including accepted

Explore the historical Whois records related to plasma-simulation-code.net (Domain). Get insights into ownership history and changes over time.

Explore the historical Whois records related to pop-code.net (Domain). Get insights into ownership history and changes over time.

LLM vs Human Code Dataset

A Benchmark Dataset for AI-generated and Human-written Code Classification

  Description

This dataset contains code samples generated by various Large Language Models (LLMs), including CodeStral (Mistral AI), Gemini (Google DeepMind), and CodeLLaMA (Meta), along with human-written codes from CodeNet. The dataset is designed to support research on distinguishing LLM-generated code from human-written code.

  Dataset Structure





  1.… See the full description on the dataset page: https://huggingface.co/datasets/basakdemirok/AIGCodeSet.