issue #2 Training classification model for winequality.csv dataset

[msmelo]

No description provided.

[mlopatka]

Thank you for this PR. It looks like your first contribution to the PRESC project!
Overall, it is a good start to the classification task. I do have some changes to recommend:

• Please use the dataset that is already provided in the repo, when you load it use a relative path that reflects the directory structure of the repository so that other people can run your code locally without errors.
• The relationship between quality and recommend is completely separable, I.e. the quality is a perfect predictor of whether or not it is recommended. 100% accuracy is always suspicious :)
• Please repeat the analysis dropping quality from the features.
• The confusion matrices for every classifier considered are not adding important information. Please summarize the interesting findings of this analysis in light of different classifier performance.

[msmelo]

Hello @mlopatka
Thanks for the feedback! I've made some changes and I've deleted the csv file from the folder. I struggled a little bit to commit the changes, but I believe the file is updated now.