datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
glaive-code-assistant
Glaive Code Assistant
Glaive Code Assistant dataset formatted for training assistant models with the following prompt template:
<s>[INST] {question} [/INST] {answer} </s>
Trained model can be prompted in Llama style:
<s>[INST] {{ user_msg }} [/INST]
Code-Review-Assistant
Dataset Card for Code Review Assistant Training Dataset
Dataset Description
Overview
This is the training split of the Code Review Assistant Dataset - a comprehensive synthetic dataset designed for fine-tuning AI models in Python code review, security analysis, and code quality assessment.
Dataset Summary
Curated by: Alen Philip
Language: English (with Python code examples)
License: cc-by-nc-4.0
Total Examples: 13,670
Purpose: Training data for code… See the full description on the dataset page: https://huggingface.co/datasets/alenphilip/Code-Review-Assistant.Code-Review-Assistant-Eval
Dataset Card for Code Review Assistant Evaluation Dataset
Dataset Description
Overview
This is the evaluation split of the Code Review Assistant Dataset - a held-out set for validating and benchmarking models trained on the training dataset. Contains diverse Python code review examples for comprehensive model evaluation.
Dataset Summary
Curated by: Alen Philip
Language: English (with Python code examples)
License: cc-by-nc-4.0
Total Examples: 1,726… See the full description on the dataset page: https://huggingface.co/datasets/alenphilip/Code-Review-Assistant-Eval.