Back To Module Home

Make Your Own Neural Network in Python

0% completed

Prologue

The Search for Intelligent Machines A Nature Inspired New Golden Age

Part 1 - A Little Background

Humans vs. Computers A Simple Prediction Machine Estimate the Constant Iteratively Classify vs. Predict Build a Simple Classifier Errors in the Training Classifier Refine the Parameters of the Training Classifier Set Up a Learning Rate in the Training Classifier Limitations of Linear Classifiers Represent Boolean Functions with Linear Classification

Part 2 - Let's Get Started!

Neurons: Nature’s Computing Machines How Neurons Work Activation Functions Replicate Neurons in an Artificial Model Follow Signals through a Simple Network Calculate Neural Network Output Matrix Multiplication Calculate the Inputs for Internal Layers A Three-Layer Example: Working on the Input Layer A Three-Layer Example: Working on the Hidden Layer A Three-Layer Example: Working on the Output Layer

Part 3 - Backward Propagation of Error

Learn Weights from More than One Node Error Backpropagation from More Output Nodes Backpropagation: Splitting the Error Backpropagation: Recombine the Error Error Backpropagation with Matrix Multiplication

Part 4 - Adjusting the Link Weights

Update Weights Embrace Pessimism Gradient Descent Algorithm Transform the Output into an Error Function Use Gradient Descent to Update Weights Choose the Right Weights Iteratively The Error Slope between the Input and Hidden Layers Weight Updates Calculated Prepare Data: Inputs and Outputs Prepare Data: Random Initial Weights

Part 5 - A Gentle Start with Python

An Introduction to Python Loops Arrays Plot the Arrays Objects Methods

Part 6 - Neural Network with Python

Build the Neural Network Class Initialize the Neural Network Weights: The Heart of the Network Query the Network Apply the Sigmoid Function Python Code for a Basic Neural Network Train the Network Complete Code for a Neural Network

Part 7 - Testing Neural Network against MNIST Dataset

The MNIST Dataset MNIST Data File Get the Dataset Ready Plot the Data Points Prepare the MNIST Training Data Rescale the Target Output Create a Target Array Update the Neural Network Code Test the Network on a Subset Test the Network on the Whole Dataset Evaluate the Performance of the Neural Network

Part 8 - Some Suggested Improvements

Tweak the Learning Rate Do Multiple Runs Change the Network Shape The Final Neural Network for MNIST Dataset

Part 9 - Even More Fun!

Query on Our Own Handwriting Inside the Mind of a Neural Network More Brain Scans Create New Training Data with Rotations

Appendix: A Small Guide to Calculus

A Gentle Introduction to Calculus A Flat Line A Sloped Straight Line A Curved Line Calculus Not by Hand Calculus without Plotting Graphs Patterns Chain Rule Handling Independent Variables

Conclusion

...

/

Final Remarks

Final Remarks

Here's a quick summary of the "Fundamentals of Neural Network" module.

Congratulations! We’ve successfully completed the “Fundamentals of Neural Network” module. 🎉

Summary

...