datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
rubygems-20241031rubygems-20230301cornstack-ruby-v1
CoRNStack Ruby Dataset
The CoRNStack Dataset, accepted to ICLR 2025, is a large-scale high quality training dataset specifically for code retrieval across multiple
programming languages. This dataset comprises of <query, positive, negative> triplets used to train nomic-embed-code,
CodeRankEmbed, and CodeRankLLM.
CoRNStack Dataset Curation
Starting with the deduplicated Stackv2, we create text-code pairs from function docstrings and respective code. We filtered out… See the full description on the dataset page: https://huggingface.co/datasets/nomic-ai/cornstack-ruby-v1.cornstack_ruby_ru_enThe CoRNStack Dataset translated into Russian. Translation was done with Qwen3-8B model.
CodeSearchNet-ruby-qrelsEmploying the CoIR evaluation framework's dataset version, utilize the code below for assessment:
import coir
from coir.data_loader import get_tasks
from coir.evaluation import COIR
from coir.models import YourCustomDEModel
model_name = "intfloat/e5-base-v2"
# Load the model
model = YourCustomDEModel(model_name=model_name)
# Get tasks
#all task ["codetrans-dl","stackoverflow-qa","apps","codefeedback-mt","codefeedback-st","codetrans-contest","synthetic-
# text2sql","cosqa","codesearchnet"… See the full description on the dataset page: https://huggingface.co/datasets/CoIR-Retrieval/CodeSearchNet-ruby-qrels.CodeSearchNet-ruby-queries-corpus
Dataset Card for "CodeSearchNet-ruby-queries-corpus"
More Information needed
ml-table-retrieval-benchmarks
ML Table Retrieval Benchmarks
Benchmark data for ml-table-retrieval: queries, database schemas, and SQLite databases for evaluating table-set retrieval and reranking.
Download everything with:
pip install huggingface_hub
python -c "from huggingface_hub import snapshot_download; snapshot_download('Rubyando59/ml-table-retrieval-benchmarks', repo_type='dataset', local_dir='data')"
or, from inside the ml-table-retrieval repo, python scripts/download_data.py (pulls only the splits… See the full description on the dataset page: https://huggingface.co/datasets/Rubyando59/ml-table-retrieval-benchmarks.Stack2Graph_KG_ruby
Ruby StackOverflow Knowledge Graph
Summary
This Hugging Face dataset repository contains the Ruby shard of the Stack2Graph StackOverflow Knowledge Graph.
Hugging Face uses one dataset repository per programming language, so this repository is directly cloneable without an extra top-level archive wrapper.
The artifact is optimized for graph-based retrieval, SPARQL analytics, and retrieval-augmented question answering over Stack Overflow content.
Stack2Graph source:… See the full description on the dataset page: https://huggingface.co/datasets/Mo7art/Stack2Graph_KG_ruby.marin-starcoderdata_rubygnn-ruby-code-study
GNN Ruby Code Study
Systematic study of Graph Neural Network architectures for Ruby code complexity prediction and generation.
Paper: Graph Neural Networks for Ruby Code Complexity Prediction and Generation: A Systematic Architecture Study
Dataset
22,452 Ruby methods parsed into AST graphs with 74-dimensional node features.
Split
Samples
File
Train
19,084
dataset/train.jsonl
Validation
3,368
dataset/val.jsonl
Each JSONL record contains:
repo_name:… See the full description on the dataset page: https://huggingface.co/datasets/timlawrenz/gnn-ruby-code-study.codexglue_code2text_ruby
Dataset Card for "codexglue_code2text_ruby"
More Information needed
the-stack-ruby-clean
Dataset 1: TheStack - Ruby - Cleaned
Description: This dataset is drawn from TheStack Corpus, an open-source code dataset with over 3TB of GitHub data covering 48 programming languages. We selected a small portion of this dataset to optimize smaller language models for Ruby, a popular statically typed language.
Target Language: Ruby
Dataset Size:
Training: 900,000 files
Validation: 50,000 files
Test: 50,000 files
Preprocessing:
Selected Ruby as the target language due to its… See the full description on the dataset page: https://huggingface.co/datasets/ammarnasr/the-stack-ruby-clean.RubyCraft-3.4-Instruct
RubyCraft-3.4-Instruct
RubyCraft-3.4-Instruct is a synthetically generated and rigorously filtered dataset designed to rapidly adapt Small Language Models (SLMs) to the latest programming language specifications, specifically Ruby 3.4.
This dataset demonstrates how high-performance code specialist models can be created at a negligible cost (approximately €4.20), effectively eliminating dependency on expensive, monolithic LLM APIs.
📊 Dataset Summary
The dataset is… See the full description on the dataset page: https://huggingface.co/datasets/mehmetdavut/RubyCraft-3.4-Instruct.CodeSearchNet-ccr-ruby-queries-corpus
Dataset Card for "CodeSearchNet-ccr-ruby-queries-corpus"
More Information needed
CodeSearchNet-ccr-ruby-qrelsEmploying the CoIR evaluation framework's dataset version, utilize the code below for assessment:
import coir
from coir.data_loader import get_tasks
from coir.evaluation import COIR
from coir.models import YourCustomDEModel
model_name = "intfloat/e5-base-v2"
# Load the model
model = YourCustomDEModel(model_name=model_name)
# Get tasks
#all task ["codetrans-dl","stackoverflow-qa","apps","codefeedback-mt","codefeedback-st","codetrans-contest","synthetic-
# text2sql","cosqa","codesearchnet"… See the full description on the dataset page: https://huggingface.co/datasets/CoIR-Retrieval/CodeSearchNet-ccr-ruby-qrels.wildlife-collection4
Wildlife Multimodal3 Data Notes
Dataset summary
This repository contains a preparation pipeline and a small metadata sample for Wildlife work with Multimodal3 inputs. It does not claim to be a complete benchmark release; the loader documents how source data is normalized and validated.
Included material
clean.py — loading, cleaning, and split preparation code.
dataset_infos.json — schema and split metadata.
metadata_sample.jsonl — small… See the full description on the dataset page: https://huggingface.co/datasets/rubywhite2002/wildlife-collection4.code-text-ruby
Dataset is imported from CodeXGLUE and pre-processed using their script.
Where to find in Semeru:
The dataset can be found at /nfs/semeru/semeru_datasets/code_xglue/code-to-text/ruby in Semeru
CodeXGLUE -- Code-To-Text
Task Definition
The task is to generate natural language comments for a code, and evaluted by smoothed bleu-4 score.
Dataset
The dataset we use comes from CodeSearchNet and we filter the dataset as the following:
Remove… See the full description on the dataset page: https://huggingface.co/datasets/semeru/code-text-ruby.details_Yuma42__KangalKhan-Ruby-7B
Dataset Card for Evaluation run of Yuma42/KangalKhan-Ruby-7B
Dataset automatically created during the evaluation run of model Yuma42/KangalKhan-Ruby-7B on the Open LLM Leaderboard.
The dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 2 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_Yuma42__KangalKhan-Ruby-7B.ruby-code-instructions-80k
Ruby Code Instructions - 80K
Instruction-tuning dataset of Ruby functions or methods paired with short natural-language instructions derived from repository docstrings or inline comments.
What changed in this release
This release keeps the original public rows and format, but makes the dataset easier to use responsibly:
exact duplicate rows were removed again using normalized instruction + output hashing
deterministic train, validation, and test splits were added
the… See the full description on the dataset page: https://huggingface.co/datasets/NickIBrody/ruby-code-instructions-80k.details_Yuma42__KangalKhan-Beta-Ruby-7B
Dataset Card for Evaluation run of Yuma42/KangalKhan-Beta-Ruby-7B
Dataset automatically created during the evaluation run of model Yuma42/KangalKhan-Beta-Ruby-7B on the Open LLM Leaderboard.
The dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_Yuma42__KangalKhan-Beta-Ruby-7B.llm-redteam-corpus-taxonomy-core
LLM Red-Team Corpus + Taxonomy — Core
44,681 prompts for red-teaming and safety-testing LLMs. We collected them
from 9 public sources, checked every license, masked personal data, and saved
where each prompt came from. Load it in one line.
Use one clean dataset instead of downloading and fixing 9 red-team benchmarks by
hand. Every prompt keeps its source, license, and citation. The
core tier is safe to use in commercial products — every source is verified
MIT / Apache-2.0 /… See the full description on the dataset page: https://huggingface.co/datasets/RUBYGLASK/llm-redteam-corpus-taxonomy-core.smollm3-stack-v2-Rubyinstructional_code-search-net-ruby
Dataset Card for "instructional_code-search-net-ruby"
Dataset Summary
This is an instructional dataset for Ruby.
The dataset contains two different kind of tasks:
Given a piece of code generate a description of what it does.
Given a description generate a piece of code that fulfils the description.
Languages
The dataset is in English.
Data Splits
There are no splits.
Dataset Creation
May of 2023
Curation Rationale
This dataset… See the full description on the dataset page: https://huggingface.co/datasets/Nan-Do/instructional_code-search-net-ruby.eval-fsr-a1-stack-ruby-swe-r469-rf0710-tracesRuby-Code-Large
Ruby-Code-Large
Ruby-Code-Large is a large-scale corpus of Ruby programming language source code comprising 331,743 code samples stored in .jsonl format. The dataset is designed to support research and development in large language model (LLM) pretraining, static analysis, web application development, and software engineering automation within the Ruby ecosystem.
By offering a substantial, language-focused dataset, Ruby-Code-Large enables targeted experimentation in dynamic… See the full description on the dataset page: https://huggingface.co/datasets/ajibawa-2023/Ruby-Code-Large.fda-warning-letter-escalation
FDA Warning Letter Escalation Dataset
Version: 1.0.0 | Records: 14,810 | Price: $2,000 | Source: FDA (public domain)
Dataset Summary
The dataset's core signal — whether an inspection escalated to a Warning Letter — is validated against FDA's own severity classifications: OAI inspections escalate at 64% versus 1.7% for NAI, a 39× relationship that confirms the data reflects real regulatory behavior.
The 14,810 inspections that escalated to enforcement — every… See the full description on the dataset page: https://huggingface.co/datasets/RubyIntelligence/fda-warning-letter-escalation.exp_rpt_nemotron-ruby_10k_glm_4.7_traces_jupiterspider_dbruby-rails-ja-en
Description
This is a WIP dataset!
This dataset contains Japanese and English sentences taken from Ruby and Ruby on Rails documentation websites. It also contains some synthetic data generated by Claude (TODO: label synthetic data vs. 'real' data).
Usage
Load dataset into your project
dataset = load_dataset("morinoko-inari/ruby-rails-ja-en", split="train")
Manually split into test and training sets
For example:
dataset =… See the full description on the dataset page: https://huggingface.co/datasets/morinoko-inari/ruby-rails-ja-en.rubygems-20230301