The objective of this challenge is to forecast the number of turtles caught per week per capture site. This will help Local Ocean plan their staff schedules and budget.
For more info on the project (https://zindi.africa/competitions/sea-turtle-rescue-forecast-challenge)
The model used is a sing Catboost Regressor model. The target variable used was the daily number of turtles caught per capture site before converting it to weekly number of turtles caught per capture site as stated in the Challenge