animal-clef-2026
收藏资源简介:
AnimalCLEF26 Kaggle竞赛数据集是官方AnimalCLEF26 Kaggle竞赛数据集的HuggingFace镜像版本,该版本在保留原始数据文件的基础上,增加了Parquet元数据以优化在HuggingFace平台上的使用体验。数据集主要用于图像分类任务,具体聚焦于动物再识别(animal-re-identification)。数据组织包含训练集(train)和测试集(test)两个部分,文件存储路径遵循images/*/train/**和images/*/test/**的模式。用户可通过HuggingFace的datasets库直接加载使用。关于数据集的详细背景、具体内容、规模、字段定义及使用许可等信息,需参考原始的Kaggle竞赛页面和数据文档。
The AnimalCLEF26 Kaggle Competition dataset is a HuggingFace mirror version of the official AnimalCLEF26 Kaggle Competition dataset. This version retains the original data files while adding Parquet metadata to optimize the usage experience on the HuggingFace platform. The dataset is primarily used for image classification tasks, specifically focusing on animal re-identification. The data organization includes training (train) and test (test) sets, with file storage paths following the pattern images/*/train/** and images/*/test/** respectively. Users can directly load and use the dataset via the HuggingFace datasets library. For detailed background, specific content, scale, field definitions, and licensing information, refer to the original Kaggle competition page and data documentation.
数据集概述
- 名称:AnimalCLEF26 Kaggle Competition Dataset
- 任务类别:图像分类
- 标签:动物重识别
数据集结构
该数据集在HuggingFace上以镜像形式提供,图像按分割打包为单个zip文件,包含额外的元数据以便与HuggingFace配合使用,原始文件未经修改。数据集包含以下配置:
- 配置名称:default(默认配置)
- 数据文件:
- 训练集:
data/train.zip - 测试集:
data/test.zip
- 训练集:
加载方式
使用以下Python代码加载数据集:
from datasets import load_dataset
dataset = load_dataset("BVRA/animal-clef-2026")
print(dataset["train"][0]["image"])
文档与许可
- 文档:请参考原始Kaggle数据页面:https://www.kaggle.com/competitions/animal-clef-2026/data
- 许可:在使用或重新分发此数据集前,请审阅Kaggle竞赛规则:https://www.kaggle.com/competitions/animal-clef-2026/rules#7-competition-data




