Kaggle Tips for Feature Engineering and Selection | by Gilberto Titericz | Kaggle Days Meetup Madrid

[@gehallak660]

Thanks for the talk. Precious tips!

[@SuperMsmystery]

Awesome talk

[@javqui1102]

Excellent content, but very hard to read the slides text at the bottom-right

[@luisfernandodasilvamartine8453]

Amazing!

[@swarajshinde3950]

Slides please

[@lebesgue-integral]

Thank you very much for this talk

[@leehongkai]

Thanks for the great summary and guide!

[@jfjfcjcjchcjcjcj9947]

@18:52 In that example, suppose that the table shown is our train data, now on the test data we don't have the columns "Color" nor "Target", plus our train data is of size 100 and test data of size 200. How is this approach gonna work @24:11 when your test set is of different size, plus the merge is based on a column on the train set which is absent on the test set?

[@javqui1102]

42:07