mahdiazhari / machine-learning Goto Github PK

Total 14.25 out of 15 = 95%
Written assignment: 4.75/5

• see the solutions online

Programming assignment: 9.5/10

• very well done
• nice modular code
• your search space for parameter c could have been bigger
• a bit more in depth written analysis would be better

Total: 12.5/15
Programming assignment: 9.5/10

• always plot axes names on a plot!
• get rid of the code that you don't use, those commented out print statements shouldn't be there
• analysis ok, but some minor improvements, like:
• for overfitting those are good solutions, but do you need them at all? are you overfitting?
• you could have make a table for alphas, or even make a plot.
• very good score because of the spitted cost function, nice one!

Written assignment: 3/5

• see the correct answers at https://staff.fnwi.uva.nl/m.a.migut/AUC_ML/solutions.html