Closed Zymrael closed 3 years ago
Starting Tuesday I'm going full tilt on pytorch and DiffEq. If I write code to learn I'll learn faster.
How can I contribute? Something small that would give me confidence would be a good start. I noticed you added help wanted.
Hey @kivo360, thank you for offering to help! Join us in Slack, we'll find together a first contribution that's interesting and can help you get familiar with the codebase.
Hello, Id like to contribute to this, if that is possible.
We're a bit busy with NeurIPS at the moment, but feel free to join us on Slack - unless you've joined already (and posted on #torchdyn-dev
?)
I was wondering if there is any progress on including adaptive checkpointing?
Sounds like a really useful addition to the repo. :)
We had to pause active development due to other obligations, but there should be more coming in the third quarter of 2021, including some of the newer advances in the field.
Closing as we're jumping ahead to v1.0
with a whole host of new features (currently in dev.)
Major:
HybridNeuralDE
API: modular ODE-RNN and friendsVariationalNeuralDE
APINeuralSDE
compatibility (torchsde
)Minor:
TimeSeriesDataset
DC
andStable
normalizing flow tutorialsMiscellaneous: