Predictive model for calories burnt using different features.
The instant model has a coefficient of determination of 0.97. The model is capable of predicting the calories burnt during the exercise activity of both genders based on several features such as gender, age, height, weight, duration of exercise, heart rate. Feature engineering has been performed on features such as duration of exercise and heart rate. By converting Duration and Heart Rate to second order polynomial features (in other words, the training data will include the value and the value squared). This transformation will help account for some non-linearity in the relationship between these features and the label (Calories).