Training Neural Networks in Python
2h 5mAdvanced2022-11-09
Authors
Eduardo Corpeño
Electrical Engineer, Computer Programmer, and Teacher for 15+ years
Course details
Having a variety of great tools at your disposal isn’t helpful if you don’t know which one you really need, what each tool is useful for, and how they all work. In this course learn the inner workings of neural networks, so that you're able to work more effectively with machine learning tools. Instructor Eduardo Corpeño helps you learn by example by providing a series of exercises in Python to help you to grasp what’s going on inside. Discover how to relate parts of a biological neuron to Python elements, which allows you to make a model of the brain. Then, learn how to build and train a network, as well as create a neural network that recognizes numbers coming from a seven-segment display. Even though you'll probably work with neural networks from a software suite rather than by writing your own code, the knowledge you’ll acquire in this course can help you choose the right neural network architecture and training method for each problem you face.
Skills covered
Neural Networks and Deep LearningPythonArtificial Intelligence (AI)Programming LanguagesOpen SourceSoftware DevelopmentOne-Off
Concepts
0. Introduction
- 01 - Creating a neural network in Python
- 02 - What you should know
- 03 - Using GitHub Codespaces with this course
1. Choosing a Neural Network
- 04 - What is a neural network
- 05 - Why Python
- 06 - The many applications of machine learning
- 07 - Types of classifiers
- 08 - Types of neural networks
- 09 - Multilayer perceptrons
2. The Building Blocks of Neural Networks
- 10 - Neurons and the brain
- 11 - A simple model of a neuron
- 12 - Activation functions
- 13 - Perceptrons - A better model of a neuron
- 14 - Challenge - Finish the perceptron
- 15 - Solution - Finish the perceptron
- 16 - Logic gates
- 17 - Challenge - Logic gates with perceptrons
- 18 - Solution - Logic gates with perceptrons
3. Building Your Network
- 19 - Linear separability
- 20 - Writing the multilayer perceptron class
- 21 - Challenge - Finish the multilayer perceptron class
- 22 - Solution - Finish the multilayer perceptron class
4. Training Your Network
- 23 - The need for training
- 24 - The training process
- 25 - The error function
- 26 - Gradient descent
- 27 - The Delta rule
- 28 - The Backpropagation algorithm
- 29 - Challenge - Write your own Backpropagation method
- 30 - Solution - Write your own Backpropagation method
5. Let's Make a Segment Display Classifier
- 31 - Segment display recognition
- 32 - Challenge - Design your own SDR neural network
- 33 - Solution - Design your own SDR neural network
- 34 - Challenge - Train your own SDR neural network
- 35 - Solution - Train your own SDR neural network
- 36 - 7 to 1 network GUI demo
- 37 - 7 to 10 network GUI demo
- 38 - 7 to 7 network GUI demo
Conclusion
- 39 - Next steps
Related courses
- Neural Networks and Convolutional Neural Networks Essential Training (2018)
- Deep Learning with Python: Optimizing Deep Learning Models
- Deep Learning with Python: Convolutional Neural Networks
- PyTorch Essential Training: Working with Images
- Supervised Learning Essential Training
- Deep Learning with Python and Keras: Build a Model for Sentiment Analysis
- Deep Learning with Python: Sequence Models and Transformers
- Artificial Intelligence Foundations: Neural Networks
Related learn paths
- Advance Your Skills in Deep Learning and Neural Networks
- Fundamentals to Become a Machine Learning Engineer
- Become an AI Engineer
- Advance Your Skills in Natural Language Processing
- Advance Your Skills in AI and Machine Learning
- Moving from Data Scientist to Data Analyst
- Become a Data Analyst
- Getting Started with AI and Machine Learning