Member-only story
How I won my third bronze medal on Kaggle’s July 2021 tabular competition
This morning, upon waking, I checked my emails and was very happy to learn that I had won my third bronze medal by participating in Kaggle’s July 2021 tabular competition:-
I won the bronze medal because I used sklearn’s function, RegressorChain, on the competition question’s multi-label output. The RegressorChain function is a multi-label model that arranges regressions into a chain. Each model makes a prediction in the order specified by the chain using all of the available features provided to the model plus the predictions of models that are earlier in the chain. Regressor chains provide the user with as a way of combining a number of regressions into a single multi-target model that is capable of exploiting correlations among targets.
I have figured out that the best way to win medals in Kaggle competitions is to endeavour to use unique algorithms. It is important, therefore, to read the sklearn documentation thoroughly to try to find out of the ordinary models that will provide new ways of accomplishing machine learning. The link to the sklearn library can be found here:-https://scikit-learn.org/stable/
The link to the Jupyter Notebook that I created to win my third Bronze Medal can be found in my personal Kaggle account, being here:- https://www.kaggle.com/tracyporter/jul-21-mlpr-reg-chain