Hack_the_IBS

Inspiration

So, we decided to go with this challenge because we want to fight IBS and help people

What it does

It outputs contributions of each food to the symptom that it predicts

How we built it

We used pandas to format the data initially. Then we used sklearn to regularize and normalize our data using standard scaler and normalize function. After that we used autosklearn.regression to find suitable model for us. The best performance was achieved with KNN and decision trees

Challenges we ran into

Feature engineering

Accomplishments that we're proud of

Spearman correlation pval of 0.0012

What we learned

How to use auto_ml

What's next for Hack_the_IBS

We want to improve it even more

Built With

• autosklearn