1/12/18

Today I continued working on the artificial neural network that I discussed in my previous blog. First of all, I learned how to encode categorical data into numerical data that can be used by the artificial neural network. For example, for the categorical variable gender, "male" can be encoded as 0 and "female" can be encoded as 1 using label encoders in python. I also learned how to standardize data, which is necessary in neural networks so that all independent variables are weighed equally at first. For example, gender can be represented as a 0 or 1, but salary can be in the $10000-$200000 range, so there is a function in python that "normalizes" these data sets so that each data point for each category has a numerical value in a set range (ex -2 to 2). Finally, I learned about the classes needed to initialize the neural network and reviewed the process of neural networks learning to make sure I was on the same page as the lecturer.