How to Quickly Save a Trained Machine Learning Model
This blog post discusses how to quickly save a trained machine learning model that suggests recipes based on the user’s data, and nutritional value using Joblib and Pickle. The reason I decided to work on this is that the domain of food is varied and complex and presents many challenges to developing a recommender system. […]
Classical & Statistical Time Series Modelling of United Health Group’s Stock Price
Time series is different from a regular regression problem because it is time dependent. The basic assumption of a linear regression that the observations are independent doesn’t hold in this case. Along with an increasing or decreasing trend, most time series have some form of seasonality trends, i.e. variations specific to a particular time frame. […]
Novelty, Anomaly and Segmentation Discovery using Matrix Profile
In this notebook, novelty and anomaly and segmentation discovery using Matrix Profile. We are using Stumpy for time series data mining tasks. We’ll examine a data set containing daily opening values for the United Health Group from 2016 up to present day. UnitedHealth Group Incorporated is an American for-profit managed health care company based in Minnetonka, […]