地球资源数据云——数据资源详情

巧克力分类

发布时间:2026-03-17 15:34:24资源ID:2033809130710798337资源类型:免费

该数据集《Chocolate Classification》主要用于二分类任务,数据形态以图像为主。 题目说明:Dataset to classify between white & dark chocolates from brands like Godiva, etc 任务类型:图像二分类。 建议流程:先检查类别分布与脏样本,再用迁移学习(如 ResNet/EfficientNet)建立基线。 评估建议:使用分层切分或交叉验证,优先关注 F1、Recall、AUC 等分类指标。 可用文件:未检测到标准 CSV,可优先查看目录中的索引或说明文件。 Hi Everyone <br> I have a deep love for chocolates and I wanted to share this love with the community via a dataset. <br> I hope you have fun with it! and if you did I would really appreciate an upvote <p> <img src="https://www.googleapis.com/download/storage/v1/b/kaggle - user - content/o/inbox%2F2862643%2Fcdebf0a07a4d3ce850f966929a3c4a95%2FImage_1.jpeg?generation=1658239396446139&alt=media"> <img src="https://www.googleapis.com/download/storage/v1/b/kaggle - user - content/o/inbox%2F2862643%2F3fabbaf24d02d97cf79d1d993abcfbe0%2FImage_139.jpeg?generation=1658239430036146&alt=media"> </p> Dataset Description Around 70 images for each type of chocolate - dark and white for binary classification (Colored and mostly chocolate bars). I will be publishing a demo notebook soon.

巧克力分类

摘要概览

该数据集《Chocolate Classification》主要用于二分类任务,数据形态以图像为主。 题目说明:Dataset to classify between white & dark chocolates from brands like Godiva, etc

任务类型:图像二分类。

建议流程:先检查类别分布与脏样本,再用迁移学习(如 ResNet/EfficientNet)建立基线。

评估建议:使用分层切分或交叉验证,优先关注 F1、Recall、AUC 等分类指标。

可用文件:未检测到标准 CSV,可优先查看目录中的索引或说明文件。

Hi Everyone <br> I have a deep love for chocolates and I wanted to share this love with the community via a dataset. <br> I hope you have fun with it! and if you did I would really appreciate an upvote

<p> <img src="https://www.googleapis.com/download/storage/v1/b/kaggle - user - content/o/inbox%2F2862643%2Fcdebf0a07a4d3ce850f966929a3c4a95%2FImage_1.jpeg?generation=1658239396446139&alt=media">

<img src="https://www.googleapis.com/download/storage/v1/b/kaggle - user - content/o/inbox%2F2862643%2F3fabbaf24d02d97cf79d1d993abcfbe0%2FImage_139.jpeg?generation=1658239430036146&alt=media"> </p>

Dataset Description Around 70 images for each type of chocolate - dark and white for binary classification (Colored and mostly chocolate bars). I will be publishing a demo notebook soon.