Danijar Hafner

dreamerv3. Mastering Diverse Domains through World Models

handout. Turn Python scripts into handouts with Markdown and figures

dreamerv2. Mastering Atari with Discrete World Models

dreamer. Dream to Control: Learning Behaviors by Latent Imagination

crafter. Benchmarking the Spectrum of Agent Capabilities

daydreamer. DayDreamer: World Models for Physical Robot Learning

layered. Clean implementation of feed forward neural networks

mindpark. Testbed for deep reinforcement learning

director. Deep Hierarchical Planning from Pixels

ninjax. General Modules for JAX

elements. Building blocks for productive research

embodied. Fast reinforcement learning research

computer-game. Data-oriented voxel game engine

diamond_env. Standardized Minecraft Diamond Environment for Reinforcement Learning

portal. Fast and reliable distributed systems in Python

crafter-baselines. Docker containers of baseline agents for the Crafter environment

granular. Fast dataset format and loader

sets. Read datasets in a standard way

voxel-smoothing-2d. Orientation independent bézier smoothing of voxel grids

scope. Scalable metrics logging and analysis

teleport. Efficiently send large arrays across machines

course-machine-intelligence-2. Jupyter Notebook

npgame. Write simple games in Numpy!

dotfiles. Mac and Linux config

semantic. Python

training-py. My solutions to programming puzzles

imptools. Tools for improving Python imports

awesome-rl-envs.

bridgewalk. Visual reinforcement learning benchmark for controllability

fidjax. Frechet Inception Distance in JAX

cowherd. Partially-observed visual reinforcement learning domain

definitions. Load and validate YAML definitions against a schema

map-pdf. Generate printable PDF documents from Leaflet maps

tdmpc2. Code for "TD-MPC2: Scalable, Robust World Models for Continuous Control"

modurale. Modular real time engine for computer graphics applications

couse-ml-stanford. Programming assignments for the Stanford Machine Learning course by Andrew Ng

invoicepad. Freelancer solution covering time tracking, invoice generation and archiving

seminar-knowledge-mining. Wikimedia image classification and suggestings for article authors

minerl. Java

training-cpp. My solutions to programming puzzles

training-ml. Python

mujoco_playground. An open-source library for GPU-accelerated robot learning and sim-to-real transfer.

AMO. [RSS 2025] AMO: Adaptive Motion Optimization for Hyper-Dexterous Humanoid Whole-Body Control

KibotoVision. A version of Kiboto that allows for live video analysis of a game

learnxinyminutes-docs. Code documentation written as code! How novel and totally my idea!

PPO-RND. Random network distillation on Montezuma's Revenge and Super Mario Bros.

jumper. Platformer and puzzle solving game written in Python

tensorflow. Computation using data flow graphs for scalable machine learning

planet. Deep Planning Network: Control from pixels by latent planning with learned dynamics

bigym. Demo-Driven Mobile Bi-Manual Manipulation Benchmark.

course-ml-fuberlin. Python

tf_optimizers. Optimizers in tensorflow from scratch