Skip to content


When I switched to data science, I built my digital garden, datumorphism. I deliberately designed this digital garden as my second brain. As a result, most of the articles are fragments of knowledge and require context to understand them.

Making bricks is easy but assembling them into a house is not easy. So I have decided to use this repository to practice my house-building techniques.

I do not have a finished blueprint yet. But I have a framework in my mind: I want to consolidate some of my thoughts and learnings in a good way. However, I do not want to compile a reference book, as datumorphism already serves this purpose. I should create stories.

Open Source

This is an open source project on GitHub: emptymalei/deep-learning.

How Do I Write It

I am trying out a more "agile" method. Instead of finishing the whole project at once, I will release the book by chapter. A few thoughts on this plan:

  • Each new section should be a PR.
  • Every PR is reviewed.
  • Release on every new section.


The following is an initial design of the blueprint.

  • Tech Onboarding
    • Python, Environment, VSCode, Git, ...
    • Pytorch Lightning
  • Time Series
    • Data
      • Datasets
      • Data Generating Process (DGP)
      • Data Processing
      • Data Augmentation
      • Metrics
    • Tasks
      • Forecasting
      • Classification
      • Generation
    • Models (Focus on Deep Models)
      • AR and Variants
      • RNN, e.g., LSTM
      • DeepAR
      • Conformal Prediction
      • Transformer
      • Spatial-temporal Models, e.g., GNN
  • Energy-based Models
  • Self-supervised Learning
  • Graph Neural Networks
  • Other Topics
    • Graph Neural Networks
    • Spiking Neural Networks
    • Transformers