Sign in

Audio and Machine Learning researcher || ||

How to learn from and reduce complex and high-dimensional shapes.


We cannot visualize high-dimensional data above 3 dimensions. So what do we do when we are faced with this situation that is commonplace in nearly every Data Science application? Dimension reduction techniques like PCA often fail because there is a simple assumption to these methods: the data can be linearly reduced. However, for most types of high dimensional data, there is likely a non-linear relationship and therefore we need to maintain this shape when we reduce the dimensions. This is where Manifold Learning techniques like Isomap come into play.

Note: I will be discussing numerous other Manifold Learning techniques in…

How to level-up your visualization with the already great package and gridspec


While many Data Scientists are quite comfortable with the basics of matplotlib, or more specifically pyplot, not many know that there are numerous ways to level up your visualization skills in the package. Aesthetics aside, understanding how to leverage certain tools will make your visualizations more information-dense and compact, without sacrificing interpretability.

In this article, we will be looking at gridspec, a tool that allows us to snap graphs into place and create more complex visualizations.

Note: any images without credit are by the author.


If you are familiar with the basics of pyplot and matplotlib, then learning gridspec will…

Create environments and integrate them into Jupyter Notebooks with one command

Photo by Florian Olivo on Unsplash


For those who are in the Data Science field, automation is one of their favorite words. Automating simple tasks that often take up unnecessary amounts of time is one of the best skills to have. Not only will you be able to get 10/15 minutes of work done with one line of code, but you won’t have to worry about having to remember every step of the process as you have already developed the steps for yourself.

In this article, I will show you how to use bash scripting to automate the development of Python environments in conda and ipykernel

Reflections halfway through my third year

close-up of part of a computer keyboard
close-up of part of a computer keyboard
Photo by Chris J. Davis on Unsplash

As more and more people enter the broad field of programming, it makes me want to sit back and reflect on what I have learned as I enter my third year of coding nearly every day. Whether it be for a school project or my own interest, the act of coding has really become a fundamental part of my life, and while I had my ups and downs, I love every bit of it to this day.

I had no idea what to expect when I first started this journey a little over two years ago. I would take what…

Mimic the one simple task I am doing

Photo by el alce web on Unsplash


We all know how hard the job hunt is. While I do believe that the job hunt should get harder over time, that should not mean that high-potential candidates should be left out in the dark.

While I am sure you read dozens of articles of how people got dream jobs at FAANG companies in less than a year, I have to tell you that it likely will not happen for you too. I could lie to you, but I’d rather not.

As someone who has been applying consistently for jobs, there has been one main tip that has improved…

Interpreting pitch changes for Music Information Retrieval


Now that the idea of a spectrogram is fully understood, we want to delve deeper into various structures beyond the frequency over time. When studying the sound waves over the Mel Scale, we got a taste of this — especially with the MFCCs. While visualizing MFCCs are technically not spectrograms, the rough idea still holds.

This article will be a bit more focused towards Music Information Retrieval (MIR) however, in that we are going to study the changes in pitch over time. …

Discussing the intersection of two rich fields.

Photo by Alexey Ruban on Unsplash


For those in Data Science and Machine Learning, one of the most important decisions for their future career is which domain specialty to pursue and become an expert in. Some of the most prominent fields is Natural Language Processing (NLP,) Computer Vision, and Quantitative trading. These fields are exponentially growing and new techniques within them are being developed at what seems to be a daily rate.

However, while not the most lucrative application of Machine Learning (in terms of current profitability,) the field of Music has developed tremendous insights and remarkable progression, most of which being is due to various…


#3— You show no true understanding of the Data Science Position.

Photo by Scott Graham on Unsplash

The job hunt in this field is brutal. While it may be much easier with the right connections, we also have to be real with ourselves, we likely do not have them. If we did, it would be as easy as talking to your connections, sending in your resume, and getting the job. Yes, if you have connections it is, in fact that easy.

For those who are in the more common situation, it is well known that the best way to stand out from the pack is to have a Data Science Portfolio. This can include a GitHub account…

Learn the trade secret for handling big data

Photo by Andrew Neel on Unsplash


How big is big data, really? According to Rionaldi Chandraseta, his experience when working with big data was 128 petabytes. This amount of data is truly incomprehensible. I recommend reading his article if you have never worked with data on this scale, it motivated this article as well.

Simply put, performing an operation row by row does not suffice at the level of trillions. When working with insurmountable amounts of data, one requires certain techniques to be able to calculate various statistics, alongside powerful compute clusters. There are numerous methodologies to increase performance, but the most commonly technique used is…


Over time, it should get harder too.

Photo by Muhammad Faiz Zulkeflee on Unsplash

Good things never come easy, right? This is the mentality I’ve had during the job search. Honestly, it can be quite difficult to remind myself of this when every opportunity I seem to be so close to slips away. Staying motivated during uncertain times is its own issue completely. However, as I pile up more and more rejection emails and automated apologies, I am actually glad that the process is this difficult, and personally, I hope it gets harder.

Adam, are you serious? Do you really want your entire career to rely on something as difficult as wondering if you…

Adam Sabra

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store