What is the content of building dataset included in MATLAB?

The deep learning toolbox includes many sample data sets for shallow neural networks. One of them is the building dataset whose discription is "Building energy dataset" as mentioned in MATLAB help.

I am asking for many details about the content of this dataset. When we load it

load building_dataset

we find the inputs are 14 dimension with 3 targets for 4208 sample.

For example, what are the 14 feature of a building and what are the 3 features we are training the network to predict.

Also, I found that all the values are in the range of 0:1. Are they normalized or what?

I searched many times and found nothing.

Answer 1

[Anish Walia]

Following are the details about the Inputs and Targets:

buildingInputs - a 14x4208 matrix defining fourteen attributes for 4208

different houses.

1-10. Coded day of week, time of day

11. Temperature

12. Humidity

13. Solar strength

14. Wind

buildingTargets - a 3x4208 matrix of energy usage, to be estimated

from the inputs.

For any additional information you can run the following command in MATLAB

help building_dataset

[Osama El-Ghonimy]

Thanks for your comment.

The 14 input features are clear. Any other details about 4 outputs that discripe energy? Why 4 values for the energy which is a scaler value?

Also, I found that all the values are in the range of 0:1. Are they normalized or what?