The purpose of this project is to show the power of IBM Watson Machine Learning and AutoAI to find the best model to solve a machine learning task, without any coding necessary. This will be done by training, evaluating and comparing out of 7 different algorithm to come up with the best model, which will then be deployed online. Also, there is going to be a quick test to show the model efficiency with two new samples. The full process lasted less than 10 minutes.
This is the famous problem every Machine Learning student faces when learning about the supervised algorithm KNN classifier : the Iris Flower Multiclass.
This problem consists of a .csv balanced dataset with 150 samples, all with labeled species (setosa, versicolor, virginica) which we want to predict for new data based on 4 features : sepal lengh, sepal width, petal lengh and petal width.
You can WATCH THE PROCESS HERE
You can also find the videos in this repository.