The Viral 100-Day ML Challenge, Now in Simplified Chinese

What it does

This repo translates and extends Avik-Jain’s “100-Days-Of-ML-Code” into Simplified Chinese, offering a day-by-day curriculum that runs from data preprocessing through random forests and into deep learning with TensorFlow and Keras. Each topic pairs an infographic with Markdown or Jupyter Notebook implementations, turning the project into a hybrid study guide and code reference.

The interesting bit

The project treats the original English plan as a substrate rather than scripture: some days are pure code, others are math refreshers via 3Blue1Brown or Coursera notes, and a few are honest admissions that the author needed to catch up on linear algebra. It is less a polished course and more a public study journal with 22,000+ spectators.

Key highlights

• Covers supervised classics (regression, k-NN, SVM, decision trees, random forests) and unsupervised basics (K-means, hierarchical clustering)
• Bundles daily infographics alongside Markdown explanations and Jupyter Notebook code
• Includes Chinese-language references to external resources, from Caltech lectures to the Python Data Science Handbook
• Explicit translation guidelines and FAQ for contributors
• Mixes Scikit-Learn implementations with deep-learning primers using Keras and TensorFlow

Caveats

• Some days are consumption-only (video notes, reading) rather than hands-on code
• At least one implementation is marked “待添加代码” (pending), so completeness varies by day
• The README is a long scroll of daily diary entries; finding a specific algorithm requires patience or Ctrl+F

Verdict

Mandarin-speaking beginners who want a structured, community-validated path through classical ML will find a friendly companion here. Experienced practitioners or those seeking original material should look elsewhere.