MNIST

The MNIST database (Modified National Institute of Standards and Technology database[1]) is a large database of handwritten digits that is commonly used for training various image processing systems. It was created by "re-mixing" the samples from NIST's original datasets. 

Source - https://en.wikipedia.org/wiki/MNIST_database 

How to use it:  Libraries like Keras and TensorFlow provide easy access to the MNIST dataset.  

Python code to train and test autoencoder_mnist.keras model :

import tensorflow as tf

from tensorflow.keras import layers, models

import numpy as np

import matplotlib.pyplot as plt

model = keras.models.load_model('autoencoder_mnist.keras')

(x_train, _), (x_test, _) = tf.keras.datasets.mnist.load_data()

decoded_images = model.predict(x_test)

n = 10

plt.figure(figsize=(20,4))

for i in range(n):

    ax = plt.subplot(2, n, i+1)

    plt.imshow(x_test[i].reshape(28,28), cmap="gray")

    plt.title("Original")

    plt.gray()

    ax.axis('off')

    ax = plt.subplot(2, n, i+1+n)

    plt.imshow(decoded_images[i].reshape(28,28), cmap="gray")

    plt.title("Reconstructed")

    plt.gray()

    ax.axis('off')

plt.show()