Pandas Completion

 Working through pandas 26, 27, 28:

So what is actually going on here. As one looks at the original diabetes dataset

as a whole, it soons becomes obvious one doesn't really know how the y values

were arrived at. All we have to go on is our complex data. That is a machine learning situation.

For the fun of it, we will use 80% of the dataset to train the computer. We will then

ask it to provide y or target values for the 20% we have set aside. We will then evaluate

te computer's performance in this task.

We are asking the computer to build a linear regression model. That is the 'fit'. We

will also use linear regression to evaluate the predicted outcomes against the real

in our 20% of the data.

We can redo the exercise any number of times. We seem to be scoring with approximately

45 % accuracy every time...

 This is the end to the pandas for machine learning series. We are going back to Machine

Learning, but this time with Pytorch.

                                                             *     *     *