cifar-10-batches-py
收藏阿里云天池2026-05-15 更新2025-04-19 收录
下载链接:
https://tianchi.aliyun.com/dataset/201863
下载链接
链接失效反馈官方服务:
资源简介:
经典的计算机视觉常用数据集。CIFAR10数据集共有60000个样本,每个样本都是一张32*32像素的RGB图像(彩色图像),每个RGB图像又必定分为3个通道(R通道、G通道、B通道)。这60000个样本被分成了50000个训练样本和10000个测试样本。CIFAR10数据集是用来监督学习训练的,那么每个样本就一定都配备了一个标签值(用来区分这个样本是什么),不同类别的物体用不同的标签值,CIFAR10中有10类物体,标签值分别按照0~9来区分,他们分别是飞机( airplane )、汽车( automobile )、鸟( bird )、猫( cat )、鹿( deer )、狗( dog )、青蛙( frog )、马( horse )、船( ship )和卡车( truck )。
A classic and widely used dataset in computer vision. The CIFAR-10 dataset contains a total of 60,000 samples, where each sample is a 32×32 pixel RGB color image that necessarily consists of three channels: the R (Red) channel, G (Green) channel, and B (Blue) channel. These 60,000 samples are divided into 50,000 training samples and 10,000 test samples. As CIFAR-10 is intended for supervised learning, each sample is paired with a corresponding label value to identify its category. Different object categories use distinct label values. There are 10 object categories in total in CIFAR-10, which are distinguished by label values ranging from 0 to 9, namely airplane, automobile, bird, cat, deer, dog, frog, horse, ship, and truck.
提供机构:
阿里云天池
创建时间:
2025-04-17
搜集汇总
数据集介绍

背景与挑战
背景概述
CIFAR-10是一个经典的计算机视觉数据集,包含60000张32x32像素的RGB图像,分为50000张训练图像和10000张测试图像,涵盖10个不同类别,例如飞机、汽车和鸟类等。
以上内容由遇见数据集搜集并总结生成



