Rittik Ghosh Articles


Depolying Airflow DAGs using Google Cloud Composer

"Airflow is a platform to programmatically author, schedule and monitor workflows. Use Airflow to author workflows as Directed Acyclic Graphs (DAGs) of tasks. The Airflow scheduler executes your tasks on an array of workers while following the specified dependencies. Rich command line utilities make performing complex surgeries on DAGs a …


Using a Neural Network to predict the Australian Open

I was introduced to tennis when I was very young by my grandfather and have distinct memories of watching Sampras, Agassi, and Hingis. While I no longer avidly follow the sport, I wanted to use a neural network to predict tennis matches and see how it performs when compared with …


Simulating the english premier league with pymc3

We will use pymc3 to simulate a season of the English Premier League. The models are based on the work of Baio and Blangiardo. We are using data from the 2018-2019 season gathered from Wikipedia. In addition to this, each team was given an unique numeric identifier such as 0 …

SMOTE (Synthetic Minority Oversampling Technique)

Several machine learning classification techniques tend to perform poorly on datasets where the target class (the minority class) represents a small fraction of the overall data. However, sometimes it is the minority class that we are interested in. Examples include medical applications in which we try to predict the occurrence …


Minkowski distance and its effects on KNN Classification

Minkowski distance is a generalized version of the distance calculations we are accustomed to. It can be defined as: Euclidean & Manhattan distance: Manhattan distances are the sum of absolute differences between the Cartesian coordinates of the points in question. Manhattan distances can be thought of as the sum of the …


OpenCV thresholding meets historic works of art

OpenCV or Open source computer vision is a popular image processing library. Originally developed 20 years ago by Intel, it has found widespread adoption in computer vision applications and has over 18 million downloads. Here we explore the effects of OpenCV thresholding on historical artworks. Thresholding is exactly what it …