annotated_deep_learning_paper_implementations
Free🧑🏫 60+ Implementations/tutorials of deep learning papers with side-by-side notes 📝; including transformers (original, xl, switch, feedback, vit, ...), optimizers (adam, adabelief, sophia, ...), ga
About annotated_deep_learning_paper_implementations
This is a collection of simple PyTorch implementations of neural networks and related algorithms, documented with explanations that render as side-by-side formatted notes. The project covers over 60 implementations including Transformers (original, XL, Switch, Feedback, Vision, etc.), diffusion models (DDPM, DDIM, Latent Diffusion, Stable Diffusion), GANs (CycleGAN, StyleGAN2), reinforcement learning (PPO, DQN), optimizers (Adam, AdaBelief, Sophia-G), normalization layers, capsule networks, graph neural networks, and more. It is actively maintained and aims to help users understand deep learning paper algorithms better.
Key Features
Pros & Cons
- Well-documented with detailed explanations alongside code
- Covers a broad range of topics (vision, NLP, RL, optimization, etc.)
- All implementations are in PyTorch, a popular framework
- Free and open source with permissive license
- Continuously updated with new papers and techniques
- Only provides PyTorch implementations (no TensorFlow/JAX versions for most)
- May not cover every recent paper; focuses on selected influential works
- Requires familiarity with PyTorch and deep learning concepts
- Limited interactive features; primarily a static website and code repository
Best For
Alternatives to annotated_deep_learning_paper_implementations
d2l-zh
《动手学深度学习》：面向中文读者、能运行、可讨论。中英文版被70多个国家的500多所大学用于教学。
keras
Deep Learning for humans
face_recognition
The world's simplest facial recognition api for Python and the command line
cs-video-courses
List of Computer Science courses with video lectures.
ML-For-Beginners
12 weeks, 26 lessons, 52 quizzes, classic Machine Learning for all
awesome-scalability
The Patterns of Scalable, Reliable, and Performant Large-Scale Systems