Fashion-MNIST is a dataset of Zalando's article images—consisting of a training set of 60,000 examples and a test set of 10,000 examples. Each example is a 28x28 grayscale image, associated with a label from 10 classes. We intend Fashion-MNIST to serve as a direct drop-in replacement for the original MNIST dataset for benchmarking machine learning algorithms. It shares the same image size and structure of training and testing splits.

Instruction

Get the Data

• Labels

Each training and test example is assigned to one of the following labels:

Usage

• Loading data with Python (requires NumPy)

Use utils/mnist_reader in this repo:

import mnist_reader
X_train, y_train = mnist_reader.load_mnist('data/fashion', kind='train')
X_test, y_test = mnist_reader.load_mnist('data/fashion', kind='t10k')

• Loading data with Tensorflow

Make sure you have downloaded the dataand placed it in data/fashion. Otherwise, Tensorflow will download and use the original MNIST.

from tensorflow.examples.tutorials.mnist import input_data
data = input_data.read_data_sets('data/fashion')

data.train.next_batch(BATCH_SIZE)

Note, Tensorflow supports passing in a source url to the read_data_sets. You may use:

data = input_data.read_data_sets('data/fashion', source_url='http://fashion-mnist.s3-website.eu-central-1.amazonaws.com/')

Also, an official Tensorflow tutorial of using tf.keras, a high-level API to train Fashion-MNISTcan be found here.

• Loading data with other machine learning libraries

To date, the following libraries have included Fashion-MNIST as a built-in dataset. Therefore, you don't need to download Fashion-MNIST by yourself. Just follow their API and you are ready to go.

1. Apache MXNet Gluon

2. deeplearn.js

3. Kaggle

4. Pytorch

5. Keras

6. Edward

7. Tensorflow

8. Torch

9. JuliaML

10. Chainer

You are welcome to make pull requests to other open-source machine learning packages, improving their support to Fashion-MNIST dataset.

• Loading data with other languages

As one of the Machine Learning community's most popular datasets, MNIST has inspired people to implement loaders in many different languages. You can use these loaders with the Fashion-MNIST dataset as well. (Note: may require decompressing first.) To date, we haven't yet tested all of these loaders with Fashion-MNIST.

1. C

2. C++

3. Java

4. Pythonand this and this

5. Scala

6. Go

7. C#

8. NodeJS and this

9. Swift

10. R and this

11. Matlab

12. Ruby

Citation

Please use the following citation when referencing the dataset:

@online{xiao2017/online,
  author       = {Han Xiao and Kashif Rasul and Roland Vollgraf},
  title        = {Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms},
  date         = {2017-08-28},
  year         = {2017},
  eprintclass  = {cs.LG},
  eprinttype   = {arXiv},
  eprint       = {cs.LG/1708.07747},
}