Start from a pre-trained model for Proxima run

globus-labs / cascade

Fast and reliable dynamic simulations by concurrently learning, auditing, and using surrogates

MIT License

1 stars 0 forks source link

Closed WardLT closed 4 months ago

WardLT commented 4 months ago

Avoid lengthy training cycles by starting with a model with decent weights.