Introduction to artificial neural networks, with the implementation of a multilayer perceptron.
uv venv --python 3.12 .venv && uv pip install -e .uv run python examples/mlp.pyuv python list
uv pip list
du -sh .venv
watch -n0.1 nvidia-smi
watch -n0.1 rocm-smi
uv run python -m http.server 3635 --bind 0.0.0.0
http://machine_ip:3635my intuition is that giving 80% of the dataset in one pass of our model will make it learn so fast. it does not need to reajust much its weight.