Hey there,
Hoping you are having a great day. If not lets make it worse, I am here with a new post.
As we talked about before, machine-learning uses input data called the Features and output data called the Labels to learn the model algorithm from.
So, first of all, we need lots of examples to train a neural network to recognize articles of clothing. Remember, an example is a feature label pair that we feed to the training loop.
In this case, the feature would be the input image and the label would be the correct output that specifies the piece of clothing the image depicts. Fortunately, such a dataset already exists. It's called the Fashion MNIST dataset.
Welcome to the world or the Fashion-MNIST dataset, which consists of 28 by 28 pixel gray-scale images of clothing. It contains images of t-shirts and tops, sandals, and even ankle boots.
In fact, here's a full list of all the 10 different items of clothing Fashion-MNIST contains.
Given an input image, these are our possible label outputs.
In total, the Fashion-MNIST dataset contains 70,000 images which is plenty for us to get started with.
Out of these 70,000 images, we'll use 60,000 to train the neural network. Then, we will use the remaining 10,000 images to test how well our neural network can recognize the items of clothing.
Remember, each image is 28 by 28 gray-scale pixels, so each image is 784 bytes. So our job is to create a neural network that takes the 784 bytes as input, and then identifies which of the 10 different items of clothing the image represents.
Lets make our neural network . Jump in.
Top comments (0)