How Well Can You Kaggle with Just One Hour a Day? | Hacker News Reader

How Well Can You Kaggle with Just One Hour a Day?

wandb.ai

88 pointsRafelMri15 comments

If, like me, you don't machine learning, this might be useful [1]:

Kaggle, a subsidiary of Google LLC, is an online community of data scientists and machine learning practitioners. Kaggle allows users to find and publish data sets, explore and build models in a web-based data-science environment, work with other data scientists and machine learning engineers, and enter competitions to solve data science challenges.

I did not see the word explained in the article, which was otherwise really well-written and instructive.

[1]: https://en.wikipedia.org/wiki/Kaggle

For the first few paragraphs, it went so far out of its way to explain what Kaggle is that I thought it was satire.

The massively-over-used Drake graphic was what tripped it for me. It's like something your Dad would put in a blog post, so this must be satire, right? Thanks, GP post, for the explanation.

nothing wrong with Drake. he's one of the greatest artist of our generation

No one said there's anything wrong with Drake. Using memes from five years ago that even your grandmother now posts on Facebook, well...

> I should have allowed myself the flexibility of 7 hours a week to be used as I saw fit versus one hour per day each week. Creativity and motivation comes in bursts and there were many days when I felt I wanted to go on for longer but couldn't because of the rules I set for myself. Conversely, there were days when I was exhausted and didn't have the same drive to work on the competition.

Given the title, I was expecting the opposite conclusion.

Successful writers often seem to follow a strict schedule regardless of motivation. I wonder if this helps stop burnout.

I'm definately of the "just keep going once I start" mindset but not sure how much this is because I can't force myself to work while unmotivated and if things would be better overall if I did space it out to an hour a day.

I think you're ignoring your own passive thought processes as the mind turns a problem over.

For me, I've noticed that my mind cogitates while I appear to be procrastinating.

Once there's a potential solution in sight the "keep going once started" mode activates and it's easy to keep going because you've already anticipated pain points.

From what I understand, people use a strict schedule regardless of motivation for two reasons:

- be sure that even if inspiration never strikes, you still have something after a week, a month, a year

- maximize your chances of inspiration striking

I've often heard that the first 20 minutes of doing a task are the hardest, and that's where you should force yourself the most. After that it gets easier. And if there is a "wave" of inspiration, you should ride it while it's there.

That sounds like the best of both worlds. There's always going to be an activation cost to pay, even for the most motivated.

So if I'm inspired, it'd still be helpful to put hands on keyboard and start, otherwise that inspiration might be squandered.

On the other hand, if I'm not inspired and don't become inspired in the first X minutes, there was probably a better chance of becoming inspired because I did start, but also a ceiling on X that indicates it just isn't happening today.

So maximal productiveness = {start regularly, to take advantage of inspiration} + {start regularly, to see if inspiration strikes}

And also worth differentiating forward progress tasks from epiphany tasks. Sometimes a great book still needs 2 more bad chapters to be complete, but a bad modification on a mediocre ML model doesn't help it score better.

The first 20 minutes are also true when I exercise as well. Once I'm past 15-20 minutes I find I can keep going.

Top 12% sure is impressive, but I'd caveat that this probably isn't very representative for the average person asking themselves that question. This seems to have been done by someone who already had experience with ML projects and with Kaggle specifically, and who just decided to challenge themself with this additional constraint. Also, the average person looking for a side project might not have the perseverance to stick with a challenge like this for 60 days, but might rather be tempted to pour in a lot of hours over a few long nights and weekends.

So my learning from this is: yeah, you can get a respectable Kaggle result with one hour a day, but only if you're already a semi-professional Kaggler to begin with.

The "top x%" in kaggle competition can be a bit misleading, many competitors just download some data and make 1 or 2 simple submissions, putting in couple of hours of work. Near the top of rankings when deadlines loom, many are working full time (at least if the prize money is good enough), or if the prize money is high, you will have teams of people working full time..

I hear it's really good for your pelvic floor.

Yes, I misinterpreted it as Kegel and had a very confusing first read of the article.

TL;DR: Use FastAI and save most of your time.

Reply on news.ycombinator.com