Synora Documentation#
Synora is a modular PyTorch library for world models, latent-dynamics planning, and representation learning. Train Dreamer, JEPA, IRIS, DiT, Genie, and DIAMOND agents with a unified API.
import synora
# Runs on ``pip install synora[gym]``.
agent = synora.create_model(
"dreamer", env="Pendulum-v1", env_backend="gym", total_steps=5_000
)
agent.train()
Get Started
User Guides
- Public API Quick Reference
- Training Guide
- Inference Guide
- Efficient Inference and Deployment
- Evaluation Guide
- Memory & Replay Buffers
- Environments Guide
- Environment Backends
- NuPlan Dataset
- Synora CLI
- Package Overview
- Controllers and Policies
- World Models Study Guide
- Tutorial: Plug Synora world models into RL libraries
- Modular RSSM
- Vision Components
- Datasets
- Loss Functions
- Plugin Registry
- World Models Deep Dive (Ha & Schmidhuber, 2018)
Algorithms
Reference