Finished the vectorization rewrite. #1

Merged
hjonkh merged 1 commit from vectorize into main 2026-10-06 17:39:41 +00:00
Owner
  • Nodes are now represented as weights, biases, outputs, and deltas in each layer.
  • Made Layers template classes, so I can mess around with using lower or higher precision floating point numbers, or eventually integers for quantization.
  • Removed Node.hpp, and layer.cpp as it had to be rolled into the header due to how c++ handles templates.
- Nodes are now represented as weights, biases, outputs, and deltas in each layer. - Made Layers template classes, so I can mess around with using lower or higher precision floating point numbers, or eventually integers for quantization. - Removed Node.hpp, and layer.cpp as it had to be rolled into the header due to how c++ handles templates.
hjonkh self-assigned this 2026-10-06 17:39:33 +00:00
- Nodes are now represented as weights, biases, outputs, and deltas in each layer.
- Made Layers template classes, so I can mess around with using lower or higher precision floating point numbers, or eventually integers for quantization.
- Removed Node.hpp, and layer.cpp as it had to be rolled into the header due to how c++ handles templates.
hjonkh merged commit 25faa333f3 into main 2026-10-06 17:39:41 +00:00
Sign in to join this conversation.
No reviewers
No labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
hjonkh/neural-network!1
No description provided.