🔨Building_Micrograd_from_Scratch

• The underlying principles of implementing neural network backpropagation, teachings sourced from Andrej.
• An “under the hood” knowledge of deep learning: layer details, loss functions, optimization, etc.

Project Overview

🎲Derivatives in Micrograd

• Understand the concept of derivative
• Implement function to calculate derivatives

📦Value Object of Micrograd

• Implement basic operations: sum, substract, multiply, division
• Computre gradients
• Implement of Chain rule in backpropagation

🌐Neural Network Architecture

• Build Neurons, Layers, and Multi Layer Perceptron
• Create Activation functions
• Establish forward and backward function
• Construct parameters for gradient update

💡 Using Colab to open can see the directory structure more clearly.