# Computational Graphs in Deep Learning

Imagine you are trying to solve a math problem. To solve it step by step, you would break it down, right? You’d start with simple calculations, moving from one step to the next. This is similar to what happens in deep learning. In deep learning, we use *computational graphs* to break down complex operations into simpler ones. This blog post will help you understand what computational graphs are, how they work, and why they are important in deep learning, even if you’re completely new to this topic. By the end, you’ll have a basic grasp of how computers use these graphs to learn.

# What Are Computational Graphs?

At its core, a computational graph is like a map that shows how various calculations are connected. In deep learning, we use computational graphs to represent mathematical operations that machines need to perform in order to learn from data. Think of it as a flowchart for a math problem where each point, called a *node*, represents an operation (like addition or multiplication), and the lines between them show how the results flow from one step to the next.

**Example**:

Let’s say we want to calculate:`z = (x + y) * w`

A computational graph would break this down into smaller steps, showing:

**Node 1**: First, add`x`

and`y`

**Node 2**: Then multiply the result by`w`