visitor@cyantusk.neocities.org:~$ loadinfo --project neural-net
Loading page information from: Neural Net... Done!
Neural Net
Group: João Gubert and William Weber Berrutti
Technologies used: Python, Linux
Status: Finished
Description
It's the second work (of two) developed for the college's subject Machine Learning, consisting of implementing a neural network from the ground up.
Challenges and Learnings
- Backpropagation: it was challenging to do the calculations, but we achieved it.
- Vectorization: we should've used this technique since the beginning. It would make coding execution faster and its implementation easier.
- Algorithmic optimizations: we should have done them to make the model learn faster.
- Datasets: we would like to have used more datasets to see how the program would behave.