demo_nn.py

This demo trains a two-layer neural network on XOR.

Run

python demos/demo_nn.py --epochs 1000 --lr 0.5 --hidden-size 4

Quiet smoke run (useful for CI):

python demos/demo_nn.py --epochs 20 --log-interval 20 --quiet

CLI options

Option	Default	Description
--epochs	1000	Training epochs
--lr	0.5	Learning rate
--hidden-size	4	Hidden layer width
--log-interval	100	Print every N epochs
--seed	42	Random seed
--save-graph	None	Save final prediction graph to <name>.dot
--quiet	False	Suppress console logs

Output

• Training-loss progression (unless --quiet)
• Final XOR predictions for 4 input pairs
• Optional DOT graph file when --save-graph is used