Despite recent exciting progress in Deep Learning around NLP, I want to show how simple classifiers can achieve moderate accuracy. This is a good thing because training these classifiers dose not require lots of data and computations, which is the case in training deep neural networks.
Things to be covered: data pre-processing, data-visualization.
You can click the link at the bottom of this page to see the full code in the Jupyter Notebook.
We explore the data about death rates from the five leading causes in America.
A report from the dataset page says as follows…
I am going to describe 4 steps for performing Unsupervised Learning.
(Things to be covered: data exploration, data preprocessing, PCA, k-means, t-SNE)
Famous Iris dataset will be used.
The dataset contains 150 samples, each…