Deep learning for complex dynamical systems
Torchdyn is maintained by DiffeqML, an open research group for the intersection of deep learning and dynamical systems
- pip install torchdyn === 1.0.1
Featured projects
Torchdyn-related research and implementations
Syntensor applies Torchdyn in biological systems simulation to predict and explain drug efficacy
Syntensor is implementing Torchdyn combined with methods in geometric deep learning, to model biological flux dynamically, systemically and at massive scale. Their platform generates multi-scale causal inferences for drug discovery and development.
Neural Hybrid Automata: Learning dynamics with multiple modes and stochastic transitions
Neural Hybrid Automata provide a systematic inference method based on normalizing flows, neural differential equations, and self-supervision. This is a showcase of NHAs on several tasks, including end-to-end learning of hierarchical robot controllers.
Get featured
Share your project on SlackMore from DiffeqML
Research
Try this
Neural differential equations made easy
from torchdyn.core import NeuralODE
# your preferred torch.nn.Module here
f = nn.Sequential(nn.Conv2d(1, 32, 3),
nn.Softplus(),
nn.Conv2d(32, 1, 3)
)
nde = NeuralODE(f)
...and you have a trainable model
Feel free to combine similar torchdyn classes with any PyTorch modules to build composite models. We offer additional tools to build custom neural differential equations and implicit models, including a functional API for numerical methods. There is much more in Torchdyn other than NeuralODE and NeuralSDE classes: tutorials, a functional API to a variety of GPU-compatible numerical methods, benchmarks...
Contribute to the library with your benchmark and model variants! No need to reinvent the wheel :)