# Exercise 1 - K-Nearest Neighbor Classifier for MNIST
%% Cell type:markdown id: tags:
In this exercise, we'll apply KNN Classifiers to the MNIST dataset. The aim of the exercise is to get acquainted with the MNIST dataset.
This guide uses [tf.keras](https://www.tensorflow.org/guide/keras), a high-level API to build and train models in TensorFlow.
%% Cell type:markdown id: tags:
## Install and import dependencies
We'll need [TensorFlow Datasets](https://www.tensorflow.org/datasets/), an API that simplifies downloading and accessing datasets, and provides several sample datasets to work with. We're also using a few helper libraries.
%% Cell type:code id: tags:
``` python
!pipinstall-Utensorflow_datasets
```
%%%% Output: stream
Requirement already up-to-date: tensorflow_datasets in /opt/conda/lib/python3.7/site-packages (4.6.0)
This guide uses the [MNIST](http://yann.lecun.com/exdb/mnist/) dataset—often used as the "Hello, World" of machine learning programs for computer vision. The MNIST dataset contains images of handwritten digits (0, 1, 2, etc)
We will use 60,000 images to train the network and 10,000 images to evaluate how accurately the network learned to classify images. You can access the MNIST directly from TensorFlow, using the [Datasets](https://www.tensorflow.org/datasets) API:
[1mDownloading and preparing dataset 11.06 MiB (download: 11.06 MiB, generated: 21.00 MiB, total: 32.06 MiB) to ~/tensorflow_datasets/mnist/3.0.1...[0m
[1mDataset mnist downloaded and prepared to ~/tensorflow_datasets/mnist/3.0.1. Subsequent calls will reuse this data.[0m
%% Cell type:markdown id: tags:
Loading the dataset returns metadata as well as a *training dataset* and *test dataset*.
* The model is trained using `train_dataset`.
* The model is tested against `test_dataset`.
The images are 28 $\times$ 28 arrays, with pixel values in the range `[0, 255]`. The *labels* are an array of integers, in the range `[0, 9]`. These correspond to the handwritten numbers.
Each image is mapped to a single label. Since the *class names* are not included with the dataset, store them here to use later when plotting the images:
Let's explore the format of the dataset before training the model. The following shows there are 60,000 images in the training set, and 10000 images in the test set:
Display the first 25 images from the *test set* and display the class name below each image. Verify that the data is in the correct format and we're ready to build and train the network.
If numbers are not your thing then use the [Fashion MNIST](https://github.com/zalandoresearch/fashion-mnist) dataset, which contains 70,000 grayscale images in 10 categories. The images show individual articles of clothing at low resolution (28 $\times$ 28 pixels), as seen here:
<b>Figure 1.</b><ahref="https://github.com/zalandoresearch/fashion-mnist">Fashion-MNIST samples</a> (by Zalando, MIT License).<br/>
</td></tr>
</table>
You may use Fashion MNIST for variety, and because it's a slightly more challenging problem than regular MNIST. Both datasets are relatively small and are used to verify that an algorithm works as expected. They're good starting points to test and debug code.
We will use 60,000 images to train the network and 10,000 images to evaluate how accurately the network learned to classify images. You can access the Fashion MNIST directly from TensorFlow, using the [Datasets](https://www.tensorflow.org/datasets) API:
[1mDownloading and preparing dataset 29.45 MiB (download: 29.45 MiB, generated: 36.42 MiB, total: 65.87 MiB) to ~/tensorflow_datasets/fashion_mnist/3.0.1...[0m
[1mDataset fashion_mnist downloaded and prepared to ~/tensorflow_datasets/fashion_mnist/3.0.1. Subsequent calls will reuse this data.[0m
%% Cell type:markdown id: tags:
Loading the dataset returns metadata as well as a *training dataset* and *test dataset*.
* The model is trained using `train_dataset`.
* The model is tested against `test_dataset`.
The images are 28 $\times$ 28 arrays, with pixel values in the range `[0, 255]`. The *labels* are an array of integers, in the range `[0, 9]`. These correspond to the *class* of clothing the image represents:
<table>
<tr>
<th>Label</th>
<th>Class</th>
</tr>
<tr>
<td>0</td>
<td>T-shirt/top</td>
</tr>
<tr>
<td>1</td>
<td>Trouser</td>
</tr>
<tr>
<td>2</td>
<td>Pullover</td>
</tr>
<tr>
<td>3</td>
<td>Dress</td>
</tr>
<tr>
<td>4</td>
<td>Coat</td>
</tr>
<tr>
<td>5</td>
<td>Sandal</td>
</tr>
<tr>
<td>6</td>
<td>Shirt</td>
</tr>
<tr>
<td>7</td>
<td>Sneaker</td>
</tr>
<tr>
<td>8</td>
<td>Bag</td>
</tr>
<tr>
<td>9</td>
<td>Ankle boot</td>
</tr>
</table>
Each image is mapped to a single label. Since the *class names* are not included with the dataset, store them here to use later when plotting the images:
Let's explore the format of the dataset before training the model. The following shows there are 60,000 images in the training set, and 10000 images in the test set:
Display the first 25 images from the *test set* and display the class name below each image. Verify that the data is in the correct format and we're ready to build and train the network.
