Daniel Voigt Godoy - Deep Learning with PyTorch Step-by-Step A Beginner’s Guide-leanpub

It is also possible to think of it as the simplest neural network possible: one input,
one output, and no activation function (that is, linear).
Figure 1.1 - The simplest of all neural networks
If you have read Chapter 0, you can either choose to skip to the
"Linear Regression in Numpy" section or to use the next two
sections as a review.
Data Generation
Let’s start generating some synthetic data. We start with a vector of 100 (N) points
for our feature (x) and create our labels (y) using b = 1, w = 2, and some Gaussian
noise [41] (epsilon).
Synthetic Data Generation
Data Generation
1 true_b = 1
2 true_w = 2
3 N = 100
4
5 # Data Generation
6 np.random.seed(42)
7 x = np.random.rand(N, 1)
8 epsilon = (.1 * np.random.randn(N, 1))
9 y = true_b + true_w * x + epsilon
Next, let’s split our synthetic data into train and validation sets, shuffling the array
of indices and using the first 80 shuffled points for training.
62 | Chapter 1: A Simple Regression Problem