Calories burned prediction

In this Predection, I applied different regression techniques to predict the data of calories burned in the exercises from the knowledge of duration (t), heart rate (h), body temperature (T), age (A), weight (W), height (H) and gender (G). It was verified that the variable that most influences the burning of calories is duration, followed by heart rate and body temperature. However, only with the considerations of all set of features that was possible to reduce the error to less than 1% of the mean. From the following table, we can see the models adopted an the mean error of each one.