Welcome to tinyml’s documentation!

tinyml is a small framework for writing neural networks for educational purposes. It is written in pure Python with some help from third party libraries, such as numpy, tqdm, etc. I hope this is a good source for learning neural networks and deep learning.

There might be some errors and deficiencies, and these are all my fault. If you meet any of them, please kindly leave your comments at GitHub Issue Tracker. If you are not comfortable with publicly accessible issues, please drop an email to askxzyao@gmail.com.

Contents:

• Overview
• Preliminaries
• Notations
• Theoretical Induction
  • Linear Transformation of Matrices
  • Fully Connected Layers
    • Example
  • ReLu
  • Dropout Layer
  • Softmax
  • Max Pooling Layer
  • Max Unpooling Layer
  • Convolutional Layer
    • Example
  • Transposed Convolutional Layer
  • Mean Square Loss
  • Cross Entropy Loss
• Examples
  • Back Propagation
  • The Hand Written Digits Recognition
  • The Cat Classifier
• API Reference
  • Layers
  • Optimizers
  • Loss Functions
  • Net

Indices and tables

• Index

• Module Index

• Search Page