Improving Deep Neural Networks – Ng’s 2nd Course [Nov. 2017]

For my review of the first course, see this post.

In that post I was saying that I’m flirting with the thought of paying my way through the specialization ($49 a months) after the 7-day free trial. And that’s exactly what I did.

Then I enrolled in the second course of this 5-course Deep Learning specialization and it took about 5-6 days to finish it, at a semi-relaxed pace. My final grade was 100%.

After these first two courses, I sort-of have a strategy which seems to be convenient for me. I try to watch all the lectures of a specific week in one or two sittings. This helps me achieve a good overall perspective of the material and it’s also very helpful when I do the practice quizzes. Now, a few words about the course…

Ng’s Neural Networks and Deep Learning – Some Thoughts [Nov. 2017]

I stopped taking courses on Coursera circa 2016 because I grew dispassionate about their payment models and because they stopped providing statements of accomplishment at the end of the course.

Ironically, the same payment model seems to have reignited my active interest in Coursera. About a few days ago I entered a 7-day free trial in the Deep Learning AI Specialization.

It took about 5 days to finish the first course Neural Networks and Deep Learning taught by Andrew Ng, adjunct professor at Stanford University and co-founder of Coursera. The course is 4 weeks long.

Given the level of complexity of the course, it would have taken me at least a month to get through the lectures, quizzes and programming projects, if there were no-strings attached. I take pride of the fact that I actively motivated myself to spend a couple of hours everyday working through the materials.

In the following lines I’m going to share a few thoughts after taking this first course of the specialization.

Artificial Neural Networks with Python – [New Series]

I haven’t been posting a lot of stuff on the blog lately, but I have been quite active online. I’m posting almost daily on my steemit blog and I also work on videos for my Youtube channel. Some of the videos are just stuff from my life related to nutrition and fitness while, others are programming videos.

My Steemit Crypto Experience – Personal Reflections [One Year Later]

It was about a year ago (circa July 26, 2016) when I decided to join steemit.com, a social platform that rewards its users for posting, commenting, voting, and curating content.

You can think of it like a facebook that rewards you with crypto-currency for your participation.

Atop of that, there is no central authority behind the ‘wheel’, like with conventional social media. Of course, there are games of power; but most of what happens on this platform is relatively transparent and all ‘transactions’ (transfers, withdrawals, etc.) and ‘operations’ (comments, posts, votes) can be viewed by anyone, by using different tools or by accessing the Steem blockchain programmatically (which I’ve been doing a lot).

This type of decentralized social media is likely to catch, in my view. We are averse to being controlled, but most of us are unaware of our actual ‘puppet’ status. I don’t believe in conspiracies, but I know that for-profit companies need to successfully meet their financial agendas. To get a very small sense of you being exploit, I’d recommend listening to this discussion between Sam Harris and Tristan Harris.

ACTN3 Gene and Sports Performance – A Look into 1,750 Genomes [OpenSNP]

While phenotypes are most often defined by a combination of genetic mutations (SNPs and other), there are single gene modifications that seem to have powerful phenotypic effects – think of diseases driven by single nucleotide polymorphisms. In such circumstances, you can’t do much on the ‘nurture’ side of things – when the ‘nature’ or the genetics side of it is so determining.

We’re going to take a brief look at a phenotype that seems to be strongly affected by mutations in the alpha-actin-3 or ACTN3 gene. Specifically, we take advantage of the ‘opennes’ of the OpenSNP platform where users share genetic and phenotype data.

Even more specifically, we’re looking at rs1815739 (SNP) which refers to the coding of a premature stop codon in ACTN3, which is a muscle protein located on chromosome 11. This genetic mutation seems to affect muscle performance.

Beginning Machine Learning – A few Resources [Subjective]

I’ve been meaning to write this post for a while now, because many people following the scikit-learn video tutorials and the ML group are asking for direction, as in resources for those who are just starting out.

So, I decided to put up a short and subjective list with some of the resources I’d recommend for this purpose. I’ve used some of these resources when I started out with ML. Practically, there are unlimited free resources online. You just have to search, pick something, and start putting in the work, which is probably one of the most important aspects of learning and developing any skill.

Since most of these resources involve knowledge of programming (especially Python), I am assuming you have decent skills. If you don’t, I’d suggest learning to program first. I’ll write a post about that in the future, but until then, you could start, hands-on, with the free Sololearn platform.

The following resources include, but are not limited to books, courses, lectures, posts, and Jupyter notebooks, just to name a few.

Machine Learning with Scikit-Learn – The Cancer Dataset [Work in Progress]

If you’ve been following my Youtube channel, you know I’ve been doing machine learning tutorials on the cancer dataset that comes preloaded with scikit-learn in Python.

The tutorials follow a path similar to what’s in Andreas Muller and Sarah Guido’s book on machine learning. This allows me not only to pass on knowledge to others who are interested in the details, but also to strengthen my own knowledge of these concepts.

24 video tutorials in and I realize that this is going to be a long series. I want to take the appropriate time to look into the details and to apply the very specifics for each algorithm and concept as we explore the cancer dataset.

How a Computer can Recognize an Image and Tell you what it Sees – [Desktop App]

49 lines of code. That’s all it takes to make a computer look at an image and tell you what it sees in a lifelike sounding voice.

I don’t have to reinvent the wheel. Others came before me; they made amazing creations. I can stand on their shoulders and praise their work.

To be relevant to programming, I don’t have to write a powerful algorithm for computer vision or one for speech synthesis from scratch. I can, but that’s not what I want to do with my time, especially if there are many out there that you can readily use.

What would it be like if every time you need to write code you have to program in assembler or at other lower levels of abstraction? You’d have to write thousands of lines of code for the most basic output, like a print statement for example.

Now this doesn’t mean that I’m all against that. On the contrary, I am contextually driven. I like to program my own stuff from lower levels of abstraction when I need something very specific and custom-made, which has not already been done by others.

Voice Input App in Python – Code Release and Overview [May 2017]

I built the following application, primarily, for convenience…

There are times when I don’t wanna type on my laptop. And I didn’t know of any general purpose, simple, minimal application that could do voice recognition and text input in the most basic form, as in: listen to my voice, paste what I just said, so I don’t have to type it.

I use this app in social media replies as well as when I post updates on different channels.

And, to be completely honest, there are times when I eat chocolate while watching scientific lectures (positive reinforcement). Some of these lectures spark spontaneous thoughts that I want to share, and the only ‘clean’ way to do it is by voice: one hand is used for chocolate manipulation, while the other for handling the mouse.

So, this gave me a solid reason to build this application.

Completing all 12 Programming Courses on Sololearn – [After-Thoughts]

Introduction

The world we live in today is touched by technology like never before. And the trend does not seem to decelerate any time soon. Much of technology has to do with machines powered by codes and algorithms.

Literacy and skill in programming is, therefore, a pursuit that can prevent (or delay) one from becoming obsolete in this ongoing automation. Think of robots, automated factories, self-driving trucks and autonomous drone delivery.

There are way too many online places, courses, platforms and frameworks for absolute beginners to learn programming; and most of them are free.

In this post I focus on my experience with one of them, Sololearn.

wordpress themes