Sigmoid vs ReLU Activation Functions: The Inference Cost of Losing Geometric Context

It looks like the provided code snippet for plotting decision boundaries was cut off at the end. Let's complete it and then summarize how to visualize the decision boundary plots for both Sigmoid and ReLU networks.

Completing the Decision Boundary Plot Function

The function plot_boundary is used to plot the decision boundary of a neural network on a given dataset. Here’s how you can complete this function:

python
 
1from matplotlib.colors import ListedColormap
2
3def plot_boundary(ax, net, X, y, title, color):
4    h = 0.025
5    x_min, x_max = X[:, 0].min() - 0.5, X[:, 0].max() + 0.5
6    y_min, y_max = X[:, 1].min() - 0.5, X[:, 1].max() + 0.5
7    
8    xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h))
9    grid = np.c_[xx.ravel(), yy.ravel()]
10    
11    Z = net.forward(grid).reshape(xx.shape)
12    
13    # Soft shading
14    cmap_bg = ListedColormap(["#
15
16[Read the full article at MarkTechPost](https://www.marktechpost.com/2026/04/09/sigmoid-vs-relu-activation-functions-the-inference-cost-of-losing-geometric-context/)
17
18---
19
20**Want to create content about this topic?** [Use Nemati AI tools](https://nemati.ai) to generate articles, social posts, and more.