A Python library that changes the way your approach towards Machine Vision

Image for post
Image for post
Photo by Edu Lauton on Unsplash

When I released Caer back in August of this year, I have received hundreds of emails from researchers and computer vision enthusiasts around the world thanking me for releasing the library. Their good (and bad) feedback pushed and motivated me to take the library to another level.

Today, I’m excited to announce the first-ever stable release of Caer, a lightweight open-source Python library that simplifies the way you approach Computer Vision. It abstracts away unnecessary boilerplate code enabling maximum flexibility. …


The Swiss Army Knife of Programming Languages has its problems and may be replaced by other languages better suited for the task

Image for post
Image for post
Photo by Tamara Gore on Unsplash

Ever since Python was released in the early 1990s, it has generated a lot of hype. Sure, it took the programming community at least 2 decades to appreciate its existence, but since then, it has far surpassed C, C#, Java and even Javascript in popularity.

Although Python dominates the fields of Data Science and Machine Learning, and, to some extent, Scientific and Mathematical computing, it does have its share of disadvantages when compared to newer languages like Julia, Swift and Java.

What made Python so popular?

One of the main driving points behind Python’s meteoric growth was how easy it was to learn and how powerful it was to use, making it extremely appealing to beginners and even those who shied away from programming because of the hard, unfamiliar syntax of languages like C/C++. …

And how to have a whale of a good time with it

Image for post
Image for post
`Photo by Todd Cravens on Unsplash

A couple of weeks ago, I was just about ready to release Caer, a Computer Vision library in Python to be publically available on PyPi, when I decided to send it to a friend in Alberta to tink around with it.

A few days later, I find that he’s still figuring out how to get it to work on his machine.

After dozens of hours, I finally found out why — Caer implemented code from the previous versions of other Python packages that simply weren’t available in their newer releases.

Despite having those packages installed, my friend wasn’t able to run Caer for the reasons mentioned above. …

A hackable cheatsheet to get you up and running with SQL

Image for post
Image for post
Photo by Jamie Templeton on Unsplash

I’m a Data Science enthusiast and one of the main things I deal with is Data. A lot of it.

With more than 2.5 exabytes of data generated every day, it comes as no surprise that this data needs to be stored somewhere and accessed when required.

This article presents a hackable cheatsheet to get you up and running with SQL quickly!

What is SQL?

SQL stands for Structured Query Language. It is a language for relational database management systems. SQL is used today to store, retrieve and manipulate data within relational databases.

Here’s what a basic relational database looks like.

Image for post
Image for post

Using SQL, we can interact with the database by writing queries.

A step-by-step implementation of a Dogs vs Cats classifier in Keras

Image for post
Image for post
Photo by Andrew Lancaster on Unsplash

Almost everything around you today is empowered by Machine Learning and Deep Learning. From the way Google Photos recognizes your face from a bunch of people in an image, to autonomous vehicles, these algorithms have changed the landscape of machine-centred thinking.

Among the most recent (and exciting) contributions is in the field of Computer Vision where insights are derived from media files (images and video). The most noted part is Image Classification — the ability for a machine to distinguish, or classify, between objects when given an image.

In this article, we’ll be building an image classifier to distinguish cats from dogs (and vice-versa) with sizable accuracy. …

7 Tricks and Tips to go from 50% to 90% accuracy in literally minutes!

Image for post
Image for post
Photo by SpaceX on Unsplash

So, you have gathered a dataset, built a neural network, and trained your model.

But despite the hours (and sometimes days) of work you invested to create the model, it spits out predictions with an accuracy of 50–70%. Chances are, this is not what you expected.

Here are a few strategies, or hacks, to boost your model’s performance metrics.

1. Get More Data

Deep learning models are only as powerful as the data you bring in. One of the easiest ways to increase validation accuracy is to add more data. This is especially useful if you don’t have many training instances.

If you’re working on image recognition models, you may consider increasing the diversity of your available dataset by employing data augmentation. These techniques include anything from flipping an image over an axis and adding noise to zooming in on the image. If you are a strong machine learning engineer, you could also try data augmentation with GANs. …

Included: 5 Steps to building ANY Deep Learning Model!

Image for post
Image for post
Photo by Lenin Estrada on Unsplash

For those in a hurry, here is the course:

For others:

A couple of months back, I wrote a post on how to get into Data Science. I’ve received hundreds of emails since from people all over the world, and the positive (and negative) reviews I got made me realize one thing: for new entrants to the field, Data Science is hard.

And no surprise there, Data Science simple is hard. There’s a lot of Computer Science involved besides the heavy Mathematics that comes along with it. …

Image for post
Image for post
Photo by Tim Mossholder on Unsplash

A Powerful Learning Guide for Serious Data Scientists

Data Science can be overwhelming. There are plenty who’ll tell you that in order to become a data scientist, you’ll have to learn heavy mathematics and computer science — Statistics, Calculus, Linear Algebra, Distributed Computing, Clustering and more.

It’s easy to feel like an antelope caught in the headlights — with no clue where to get started.

I understand what you’re feeling — I’ve been in your shoes not too long ago.

This post is dedicated to the resources that I wish I had at my disposal as a beginner to Data Science. I hope that sharing my experience in this post will put you on the right track to pursuing a data science career and make your learning journey that much more enjoyable. …

Image for post
Image for post
Photo by Mukil Menon on Unsplash

In Deep Learning, everyone seems to recommend using a GPU. What is it, can you do without one, and who is it exactly for?

Any data scientist or machine learning enthusiast who has been trying to elicit performance of training models at scale will at some point hit a cap and start to experience various degrees of processing lag. Tasks that take minutes with smaller training sets may now take more hours — in some cases weeks — when datasets get larger.

But what are GPUs? How do they stack up against CPUs? Do I need one for my deep learning projects?

If you’ve ever asked yourself these questions, read on…

Any data scientist or machine learning enthusiast would have heard, at least once in their life, that Deep Learning requires a lot of hardware. Some train simple deep learning models for days on their laptops (typically without GPUs) which leads to an impression that Deep Learning requires big systems to run execute. …

The bare essentials to get up and running + Helpful Resources

Image for post
Image for post
Photo by M. B. M. on Unsplash

Data Science has become a revolutionary technology that everyone seems to talk about. Hailed as the ‘sexiest job of the 21st century’, Data Science is a buzzword with very few people knowing about the technology in its true sense. While many people wish to become Data Scientists, it is essential to see the real picture.

“Data science is an interdisciplinary field that uses scientific methods, processes, algorithms and systems to extract knowledge and insights from many structural and unstructured data ” — Wikipedia

To put it simply, Data Science is a field of study and practice that’s focused on obtaining insights from data. …


Jason Dsouza

I train ConvNets. Currently building Caer, a Computer Vision library in Python.

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