【学习笔记】深度学习实战 | 流行图像数据集

简要声明

---

1. 学习相关网址
   1. [双语字幕]吴恩达深度学习deeplearning.ai
   2. Papers With Code
   3. Datasets
2. 深度学习网络基于PyTorch学习架构。
3. 本学习笔记单纯是为了能对学到的内容有更深入的理解，如果有错误的地方，恳请包容和指正。

参考文献

---

1. CIFAR-10 [https://pytorch.org/vision/stable/generated/torchvision.datasets.CIFAR10.html]
2. ImageNet [https://pytorch.org/vision/stable/generated/torchvision.datasets.ImageNet.html]
3. CIFAR-100 [https://pytorch.org/vision/stable/generated/torchvision.datasets.CIFAR100.html]
4. MNIST [https://pytorch.org/vision/stable/generated/torchvision.datasets.MNIST.html]
5. Fashion-MNIST [https://pytorch.org/vision/stable/generated/torchvision.datasets.FashionMNIST.html]

简要介绍

---

CIFAR-10

Images size	32×32×3
Label classes	10
Training set	50,000
Test set	10,000

ImageNet

Images size	224×224×3
Label classes	1000
Training set	1,281,167
Test set	100,000

CIFAR-100

Images size	32×32×3
Label classes	100
Training set	50,000
Test set	10,000

MNIST

Images size	28×28
Label classes	10
Training set	60,000
Test set	10,000

Fashion-MNIST

Images size	28×28
Label classes	10
Training set	60,000
Test set	10,000

Summary

	CIFAR-10	ImageNet	CIFAR-100	MNIST	Fashion-MNIST
Images size	32×32×3	224×224×3	32×32×3	28×28	28×28
Label classes	10	1000	100	10	10
Training set	50,000	1,281,167	50,000	60,000	60,000
Test set	10,000	100,000	10,000	10,000	10,000