cables-nl42k

# Cables Nl42K This dataset is part of the **Roboflow 100** benchmark, a diverse collection of 100 object detection datasets spanning 7 imagery domains. ## Dataset Description - **Source:** [Roboflow 100](https://github.com/roboflow/roboflow-100-benchmark) - **Category:** Real World - **License:** CC-BY-4.0 - **Format:** YOLO (LibreYOLO compatible) - **Mirrored on:** 2026-01-20 ## Dataset Statistics | Split | Images | |-------|--------| | Train | 4,816 | | Validation | 1,220 | | Test | 794 | | **Total** | **6,830** | ## Classes (11) - Antenne - BBS - BFU - Batterie - DDF - PCF - PCU AC - PCU DC - PDU - PSU - RBS ## Usage ### With LibreYOLO ```python from libreyolo import LIBREYOLO # Load a model model = LIBREYOLO(model_path="libreyoloXnano.pt") # Train on this dataset model.train(data='path/to/data.yaml', epochs=100) ``` ### Download from HuggingFace ```python from huggingface_hub import snapshot_download # Download the dataset snapshot_download( repo_id="Libre-YOLO/cables-nl42k", repo_type="dataset", local_dir="./cables-nl42k" ) ``` ## Directory Structure ``` cables-nl42k/ ├── data.yaml # Dataset configuration ├── README.md # This file ├── train/ │ ├── images/ # Training images │ └── labels/ # Training labels (YOLO format) ├── valid/ │ ├── images/ # Validation images │ └── labels/ # Validation labels └── test/ ├── images/ # Test images (if available) └── labels/ # Test labels ``` ## Label Format Labels are in YOLO format (one `.txt` file per image): ``` <class_id> <x_center> <y_center> <width> <height> ``` All coordinates are normalized to [0, 1]. ## Citation If you use this dataset, please cite the Roboflow 100 benchmark: ```bibtex @misc{rf100_2022, Author = {Floriana Ciaglia and Francesco Saverio Zuppichini and Paul Guerrie and Mark McQuade and Jacob Solawetz}, Title = {Roboflow 100: A Rich, Multi-Domain Object Detection Benchmark}, Year = {2022}, Eprint = {arXiv:2211.13523}, } ``` ## License This dataset is released under the **CC-BY-4.0** license. Please check the original source for any additional terms. ## Acknowledgments - Original dataset from [Roboflow Universe](https://universe.roboflow.com/roboflow-100/cables-nl42k) - Part of the [Roboflow 100 Benchmark](https://www.rf100.org/) - Sponsored by Intel