Predicting house prices
What we want to do is predict how much a certain house might cost, based on some variables. In order to do so you need to follow these four steps:
- Get some historical data on house sales; for this example, you'll use a dataset of about 20,000 houses in Seattle.
- Import this data to your code while applying some scaling to your variables (I'll explain scaling to you as we go).
- Build an Artificial Neural Network using any library—you'll use Keras, as it is simple and reliable.
- Train your ANN and get the results.
Now that you know the structure of your future code, you can start writing it. Since all the libraries that you'll use are available in Google Colab, you can easily use it to perform this task.
Uploading the dataset
Start by creating a new Google Colab notebook. Once we have created your new notebook, before you start coding anything, you have to upload your dataset. You can find this dataset...