Thoughts on keeping on going

I had an email today from a reader (“K”) who tried the Learn Python Challenge on Kaggle and (as does happen!) got to about Day 4 and then abandoned ship and went in search of more new stuff… The question is how to keep on going – when you get stuck, bored, de-motivated or perhaps even overwhelmed.

I’m unable to claim I have specialist insights here I’m afraid (!) but I thought I would share anyway how I have kept going so far because I think it’s a very real thing that almost everyone faces.

You don’t have to finish everything you start

In fact it may be waste of time to do so! I started my career studying music – I knew after 2 years that I was flogging the proverbial dead horse, and yet my inner critic was saying “you must finish what you started” so I wasted another 2 years on a fruitless direction!

Even my favourite data science mentor, for whom I have the greatest respect and admiration, admitted to me that he signs up for MOOC’s and seldom completes them all the way to the end! If you are learning what you need or want on a course and it’s taking you closer to the destination you have in mind then stick with it – if it’s not then perhaps it’s time to move onto something that will.

You don’t have to finish quickly

I’m busy with 2 MOOC’s at the moment which I started over 3 months ago. This seems slow but as I learn new concepts I find ways I can apply them to real projects that interest me. For example, nearly a month ago now I watched a short video from Frank Kane’s Udemy machine learning course on how to do a simple spam classifier. One month later, I haven’t gone back to the course because I’ve been tinkering with how to build a tool which will automatically log HelpDesk incidents to the right team based on how the user describes their IT problem in their own words. It’s a massive sidetrack I know, but I’ve learned and become familiar with a ton of things along the way including how to clean and manipulate messy text data, how to build scikit learn pipelines, how to work with training and test sets, how to validate results, the process of adjusting parameters to get better results, and so on. I’m still stuck video number 25 but I’ll move on in due course :).

Find something that really sparks your interest

Which brings me to my next point: from the beginning be on the lookout for small tasks or projects that interest you and which just may be within your reach given the skills you’re acquiring. When I’m working on something that I picked it holds my attention, I have the staying power, I’m engaged in the process. And one can learn more by doing than by learning sometimes!

In the beginning almost everything was beyond my reach so I picked small tasks. For example if there was something I would normally do in Excel, I’d challenge myself to do it using Python and/or Pandas instead. I also kept an eye out for small optimisation opportunities at work – programming of any kind is a time-saving gift if you have any boring repetitive tasks that you have to do: write a program to do them for you!

Commit to delivering

Another trick I’ve used is to offer to help someone (at work or an NGO even) by coding a solution for a problem or requirement they have. Once I’ve committed to producing the goods for someone there’s kind of no backing out and I’ve been forced to deliver, even when it started to feel a bit too difficult or a bit too much trouble!

Do something easy!

When you’re constantly learning new things it can be exhausting! You continually feel like you don’t know what you’re doing and you look back and fail to see the real progress you’ve made. I’m learning Pandas at the moment. In the beginning everything felt like an uphill battle: I couldn’t remember how many square brackets I needed and when to use them and where the inverted commas went and how to even extract a row of data. A couple of weeks ago I went back to have a look at those basics and I realised that I can do THAT stuff now (so something did go in!) – it’s the new concepts that feel like a struggle. That’s OK, I can deal with that!

Take a break 🙂

You don’t have to be learning 24/7 – burnout happens, especially if learning is your “after work” activity. Take a day or week off: sleep in late, watch TV, get out – just commit to a day when you’ll get back in the saddle again. A little break does wonders for the enthusiasm levels!

And finally, let me recommend to you James Clear’s wonderful article “Commit to a process, not a goal” – this spoke to me in the beginning of this journey and still does, I have it in mind nearly every day and it keeps me sane!