Bausager
fb59d61ad5
Optimazation is done
2026-08-06 18:56:36 +02:00
Bausager
7f92e6884d
Backup Save
2026-08-06 10:22:52 +02:00
Bausager
642bba1198
Softmax + Categorical Crossentropy
...
I fixed the activation+loss function, you can't select it directly, it automaticly uses it if it can. I also fixed one-hot generator. Still haven't tested the activation + loss nor any other backward function. I'll do that when I get to the optimizers which is next.
2026-08-06 09:53:03 +02:00
Bausager
fecb4c70b1
Save
...
Maybe backward on activation+loss works
2026-08-04 21:27:11 +02:00
Bausager
28c1976fb1
Save
...
Trying do make the backward to work
2026-08-04 21:05:20 +02:00
Bausager
b52c128496
Save
...
I've created the optimizer class
2026-08-04 18:55:30 +02:00
Bausager
11da534fd6
Done with activation softmax backwards
2026-07-31 19:08:57 +02:00
Bausager
d7b74ab40f
Making loss_categorical_crossentropy
2026-07-30 22:23:06 +02:00
Bausager
70e80327ef
Activation Softmax Forward done
2026-07-30 18:56:27 +02:00
Bausager
f5e0ee209b
first model
2026-07-29 17:34:22 +02:00
Bausager
c7e87fe191
model.cpp/hpp
...
I made the model struct to store layers. I'm still missing the backward functions, but it's functional
2026-07-28 19:36:26 +02:00
Bausager
189d605bfa
Dense Layer, Add
...
added the dense layer with a forward functions. I also made the add functions for most cases.
2026-06-25 19:43:07 +02:00
Bausager
cfbb06537d
Created first files for neural networks, empthy
2026-06-23 21:22:42 +02:00
Bausager