Loss does not change per epoch as the training works by trying out every node for next node always and choosing the least loss one. So, for every epoch, given the same input-output pairs, no parameters are used to find path during training and hence, it always takes the same paths for the same pairs during training, causing the same outputs and losses for every epoch.
-
Notifications
You must be signed in to change notification settings - Fork 0
A very simplistic approach to solving ARC-AGI-2. Designed to be a neural network built off of primitive functions with special tweaks. This model strives to be as simple and quick as possible as well.
Rickisterr/Primitive-Neural-Net
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
A very simplistic approach to solving ARC-AGI-2. Designed to be a neural network built off of primitive functions with special tweaks. This model strives to be as simple and quick as possible as well.
Topics
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published