Difficulty: beginner
Estimated Time: 15 minutes

This scenario is our first attempt to build one of the building blocks for the Deep Learning process. We will be using numpy library that is one of the most commonly used ones when working with arrays, vectors and tensors. It allows vectorising most of the operations making the computations more efficient.

In the next steps, you will build the simple classifying neural network and Forward Propagation flow. This is the first step when creating Deep Learning models.


You've completed Forward Propagation scenario.

You've built the simple neural network and implemented the Forward Propagation. This is the baseline for creating more complex Deep Learning models.

Forward propagation

Step 1 of 5


We will be classifying the data used in the previous scenarios when we applied Logistic Regression to the task. We will focus here only on the prediction phase, but you'll have the chance to implement the training in the later part of the course.

Below you can see the visualisation of the datasets, both linearly separated and not.

Linear dataset

Non-linear dataset

Task 1

You're going to get the data from the files. The code will be written in the forward_propagation.py. There are several help functions in the helper.py script. To load data use read_data function pointing to both data files: linear.csv and non_linear.csv.