README.md

---
license: apache-2.0
task_categories:
- text-generation
language:
- ru
- ar
- en
- zh
- de
- fr
- es
- pt
- ja
- ko
tags:
- linux
- shell
- commands
- terminal
- multilingual
- development
- system-administration
pretty_name: LinLM
size_categories:
- n<1K
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
- config_name: ru
  data_files:
  - split: train
    path: ru/train-*
dataset_info:
- config_name: default
  features:
  - name: ru
    dtype: string
  - name: ch
    dtype: string
  - name: eng
    dtype: string
  - name: de
    dtype: string
  - name: fr
    dtype: string
  - name: es
    dtype: string
  - name: pt
    dtype: string
  - name: ja
    dtype: string
  - name: ko
    dtype: string
  - name: ar
    dtype: string
  - name: completion
    dtype: string
  - name: category
    dtype: string
  splits:
  - name: train
    num_bytes: 796040
    num_examples: 2449
  download_size: 398545
  dataset_size: 796040
- config_name: ru
  features:
  - name: ru
    dtype: string
  - name: completion
    dtype: string
  - name: category
    dtype: string
  splits:
  - name: train
    num_bytes: 206046
    num_examples: 2449
  download_size: 89855
  dataset_size: 206046
---

## LinLM Dataset

A curated synthetic dataset for Linux command inference

Natural language description -> shell commands

Features: 

- Supports 10+ languages
- Arch Linux commands recognition
- Fine-tune LLM for development, system administration, file operations, Git, Docker, and more


## Usage

```python
from datasets import load_dataset

dataset = load_dataset("missvector/linux-commands")

def format_for_training(example):
    return {
        "prompt": f"Convert to Linux command: {example['eng']}",
        "completion": example['completion']
    }

training_data = dataset['train'].map(format_for_training)
```

## Out-of-Scope Use
- Not for production deployment without additional validation
- Commands should be reviewed before execution

## Related Projects

- [llama-dynamic-context](https://github.com/vifirsanova/llama-dynamic-context) - command inference tool tested with this dataset

## Citation

If you use this dataset in your research, please cite:

```bibtex
@misc{linuxcommands2025,
  author = {V. Firsanova},
  title = {LinLM Dataset},
  year = {2025},
  publisher = {Hugging Face Datasets},
  howpublished = {\url{https://huggingface.co/datasets/missvector/linux-commands}}
}
```