Adding improved_v2 to eval.py, and plot accuracy/loss curves in notebook

Readme

Adding new improved_v2 + CNN_other (does not work well) + changes to readme

Adding generate_clean_sets() in dataset.py, in order for repo github to be completly functionnal

Update readme

update changes

Update Readme

Update README.md

Adding readme. Adding improved model

Changes to train.py. Adding eval.py and plotting curves

First add of main composent of baseline. Little struggle converting baseline model from notebook to python script : might have changes. Coming next eval.py, improvements

First commit