Apply optimizer to model weights without data copy#222

Merged

milancurcic merged 13 commits intomodern-fortran:mainfrom

milancurcic:ptr

Jul 30, 2025

Commits on May 23, 2025

WIP optimizer refactor w/ pointers
milancurcic
committed

Commits on May 27, 2025

Commits on May 30, 2025

Get weights and weight gradients as 1d
milancurcic
committed

Commits on Jun 20, 2025

get_params_ptr and get_gradients_ptr for conv1d, conv2d, and locally_connected1d
milancurcic
committed
Define optimizer instance per layer to preserve memory across layers
milancurcic
committed

Commits on Jul 4, 2025

Initialization of network-wide optimizer no longer needed now that we switched to per-layer optimizer instances
milancurcic
committed

Commits on Jul 28, 2025

Bookkeeping for velocity, rms_gradient, etc.; optimizer tests now pass
milancurcic
committed

Commits on Jul 30, 2025

Remove optimizer as component to the network class
milancurcic
committed