Finally, The First Post

So, finally I'm starting to write the first post of the blog. This has been in my todo list for a while now. But as usual I have been shifting my focus to do something else, like endless cycles of checking whether the color scheme is good enough or aligning all CSS elements properly or how to write a shortcode for those repetitive block of codes et cetera et cetera.

Even today, after I started writing, I realized that I have not yet configured SEO parsing for the website yet. And started working on that. Mind you, the blog is not even live yet, let alone being it listed in a search engine!

Half an hour later, after supporting the very basic SEO stuffs, I let that go and pushed it to backlog and resumed writing. And here we are.

The General Problem by xkcdThe General Problem by xkcd

Why blog at all #

Blogging for me was always about dumping my thoughts on something in one place. I have tried this in past many times, but failed miserably as eventually I'll stop maintaining the blog. I had a blog where I used to write in Bengali during my college days. A photography blog from the same time. A relatively recent, still 5 years old, personal blog, where I planned to share insights on the work I love to do at spare time. I stopped working on all of them after some time.

However, for a long time I have planned about maintaining some sort of repository where I can log the hundreds of things I learn everyday. It can be place where I can come back later for references, if I need to do the same work again. And putting it online, so that it can help someone else looking for the same resource.

Why now of all time #

Recently I have started another blog with my wife, Little Joys of Life, hosted on WordPress. It was a personal project I took up after a long time. And while doing so I had to go through countless blogs, article, and tutorials to learn about configuring hosting services, CSS flex and grids, WordPress themes and plugins, and worst of all, PHP.

While going through those resources, I realized how easy it was for me to do something from scratch because somebody has already done it, and wrote about it. And I liked idea of doing something similar, so that someday someone else can benefit from what I have done and gone through. And believe me when I say this, something is always different. You can try following a tutorial as religiously as possible but something will not work out as it supposed to and you'll have to go down the rabbit hole to figure that out.

The final push #

In fact, that was the final push I needed to start doing this.

Recently I have started going through the fast.ai course Practical Deep Learning for Coders again. By "again", I do not mean I have finished it once. I mean I had started and dropped it after halfway through the first video.

Anyway, for that course, it is officially suggested to use some sort of cloud framework to run and train your models. However, I have a decent GPU (Nvidia 1050ti) installed in my home workstation, which I wanted to use for those times when I get maybe 15-20 minutes short spans to work on the course. Starting a cloud instance for that short period of time is really annoying and I can't let it run continuously as I'll b billed by minutes.

So I started looking for how to setup a Windows 10 machine with Nvidia GPU for that particular course, and got stuck at some point by following every article available on that in the internet. Finally it took me some tweaking here and there to get it done. It was really easy, but just not the same that other people have done. That made me realize, there might be someone else facing the exact same issue and what I have finally done might help them. So why not write it down and publish it online!

In fact, that's exactly what I'm going to write about in my next post.