Posts by Category

Episode One (IBM x UNSW machine learning project) : Data Preprocessing

Today was the first day of attending a term-long project with Data Society UNSW, in collaboration with IBM. The lecturer from Dr. Saeed Kasmani, a Senior AI ...

Episode Two (IBM x UNSW machine learning project) : Learning Models

Understanding the Roles/Links between Samples, Inference and Learning Models

I’v alway been confused of the role of statistical inference, sampling and learning models and how they all connect together. Although, I know statistical in...

Imputation : Methods

Today for my IBM project I wanted to learn more about imputation methods, selection of features, types. The following is from Flexible Imputation of Missing ...

Evaluating You’re Clustering Model

This is a continuation of my previous blog post about “Hierarchial Clustering Algorithms”. In this blog i hope to teach you teach you a bit about how to eval...

Hierarchial Clustering: Intuition and Types

Today I faced a bit of a problem on my IBM UNSW data science problem, I have the feature “project_description” which includes a range of inputs ranging from ...

Bayesian statistics vs frequentist statistics : Estimating Probability From Data

Today I was watching Kilian Weinberger lecture on Estimating Probabilities from Data: Maximum Likelihood Estimation” Cornell CS4780 SP17 and now I finally un...

Roadmap to Learn Transformers in less then 3 minutes

When I started my journey into machine learning, I remember watching an interview with Andrej Karpathy, a Sr. Director of AI at Tesla, on the Lex Fridman pod...

Functional vs Sequetial API for Keras in less then 3 minutes

Reading to Keras/Tensorflow documentation, it’s a little hard to understand what exactly is the difference between sequential and functional API and all othe...

My Machine Learing Mastery Roadmap

After watching many videos and having a solid foundational of a overview of concepts and tool in machine learing i beleive it finally time for me to make a ...

Line Search Methods for Numerical Optimisation

Linear Search Methods are a integral part of numerical optimisation, which is a set of algorithms to solve mathematics programming problems. Importantly, num...

Olympiad Level Counting Made Easy with Jensen’s inequality

Have you every wondered how high school students with no formal higher level of mathematical knowledge are able to solve complex questions like this? Prove t...

How sampling distribution is the link between Machine Learning and data science

Ever wondered how these two field like with concrete examples? Data science formull is defined “statistics, scientific computing, scientific methods, process...

How I went from 40 WPM to 100 WPM in 3 months without trying!

Flashbacks to high school when all my friends used to engage in typing test battles; touch typing, speed, the whole shebang. Then there was me, trying to loo...

My First Post

Welcome to My Blog!

CS229 Review

Review of Standford CS229 course

David Goggins : An Inspiration.

I just wanted to take some time and appreciate the man who is David Goggins. He is someone I have admired for years. The man is the definition of hard work a...

Atlassian x Alliance Datathons! Lesson I’ve Learnt

Recently I’ve had the opportunity to to participate compete in the UNSW Atlassian Allianz Data Competition at their respective headquarters! This 3-day event...

Data Viz Recap - The Good, the Bad and the Biased: Five ways Visaulisations can mislead (and how to fix them) by Danielle Szafir

Today I read a paper by Danielle Albers Szafir from University of Colorado Boulde, where she explores thew ways in which visualisation can mislead audiences....