Notice: This page requires JavaScript to function properly.
Please enable JavaScript in your browser settings or update your browser.Single Neuron Implementation | Neural Network from Scratch
Introduction to Neural Networks
course content

Course Content

Introduction to Neural Networks

Introduction to Neural Networks

1. Concept of Neural Network
What is a Neural Network?Neural Network StructureWhat is a Neuron?Activation FunctionsForward and Backward PropagationApplications of Deep Learning in the Real World
2. Neural Network from Scratch
Single Neuron ImplementationPerceptron LayersForward PropagationBackward PropagationData PreprocessingModel Training and EvaluationNeural Network with scikit-learn
3. Conclusion
Neural Networks or Traditional ModelsOther types of Neural NetworksChallenge: Manual Hyperparameter TuningChallenge: Automatic Hyperparameter TuningQuiz: ConclusionSummary

book
Single Neuron Implementation

The fundamental computational unit of a neural network is the neuron. A neuron can be visualized as a small processing unit that takes multiple inputs, processes them, and gives a single output.

Here's what happens step by step:

  1. Each input is multiplied by a corresponding weight. The weights are learnable parameters and they determine the importance of the corresponding input;
  2. All the weighted inputs are summed together;
  3. In our implementation, we will also add an additional parameter called bias to the input sum. The bias allows the neuron to shift its output up or down, adding flexibility to the modeling capability;
  4. Then the input sum is passed through an activation function. We are using the sigmoid function, which squashes values into the range (0, 1).

Note

Bias of the neuron is also a trainable parameter.

Task
test

Swipe to show code editor

Implement the basic structure of a neuron. Complete the missing parts of the neuron class:

  1. Enter the number of inputs of the neuron.
  2. Use the uniform function to generate a random bias for every neuron.
  3. Enter the activation function of the neuron.

Once you've completed this task, click the button below the code to check your solution.

Switch to desktopSwitch to desktop for real-world practiceContinue from where you are using one of the options below
Everything was clear?

How can we improve it?

Thanks for your feedback!

Section 2. Chapter 1
toggle bottom row

book
Single Neuron Implementation

The fundamental computational unit of a neural network is the neuron. A neuron can be visualized as a small processing unit that takes multiple inputs, processes them, and gives a single output.

Here's what happens step by step:

  1. Each input is multiplied by a corresponding weight. The weights are learnable parameters and they determine the importance of the corresponding input;
  2. All the weighted inputs are summed together;
  3. In our implementation, we will also add an additional parameter called bias to the input sum. The bias allows the neuron to shift its output up or down, adding flexibility to the modeling capability;
  4. Then the input sum is passed through an activation function. We are using the sigmoid function, which squashes values into the range (0, 1).

Note

Bias of the neuron is also a trainable parameter.

Task
test

Swipe to show code editor

Implement the basic structure of a neuron. Complete the missing parts of the neuron class:

  1. Enter the number of inputs of the neuron.
  2. Use the uniform function to generate a random bias for every neuron.
  3. Enter the activation function of the neuron.

Once you've completed this task, click the button below the code to check your solution.

Switch to desktopSwitch to desktop for real-world practiceContinue from where you are using one of the options below
Everything was clear?

How can we improve it?

Thanks for your feedback!

Section 2. Chapter 1
Switch to desktopSwitch to desktop for real-world practiceContinue from where you are using one of the options below
We're sorry to hear that something went wrong. What happened?
some-alt