Skeleton/Steps of typical Machine Learning model
We will follow the above steps in building all our machine learning models in this course. first and foremost we load the data, data can be from any where, local drive, our remote server or internet, our Iris data set will be from the Scikit learn server. Second step will be to extract features separately and labels separately from the data set. In our course we call our features as X and labels as Y.
Third step will be to split the X and Y into training and test data, typically we will have more data for training compared to testing. Fourth step will be to standardize the data, typically in most cases each features in a dataset will be on a different scale, we will use standard scalar in Scikit learn to standardize all our values, bring them on same scale,
Fifth step will be to choose the model for us to train and do the prediction, typically we test out with different models and choose the one that gives more accuracy in our prediction. In scikit learn when we call our Machine learning algorithm we must specify the hyperparameters for the algorithm which we can tune later for getting better accuracy.
Finally sixth and seventh steps will be to train the model using our training data and make prediction on our test data during training we can do hyperparameters tuning to improve the accuracy of our model when doing prediction on test data.
In the next article we will discuss about implementing perceptron model to make prediction on Iris data set.