TIL, 2018-05-08. Axios, recompose, thinking about ML.
Musings, JS
- GET: Don’t send data. You send things via the parameters. Search usually uses GET.
- How often do you update dependencies? Not really sure lol.
- Don’t overthink things. Just go with Bootstrap for now.
- React recompose
spinner
(withBranch
) and for lifecycle hooks. - Does this thing have to exist on every damn form? Lol.
handleSubmit = (e) => {
e.preventDefault()
}
Musings, Ruby
Array#bsearch
is really faster thanfind
, but the array must be sorted first. (Like for 50M numbers, find takes 3 seconds, and bsearch takes like 0 lol.- There was a Ruby Enterprise Edition before? (2012.)
- Basic Auth
- Admin panel: Administrate.
Musings, CS
- Learning Machine Learning:
- Studying about it and tinkering around with random data sets.
- A lot of the time, your code would not look nice. The key is, does it work?
- On math: “Let’s figure this out!”
- So you basically need linear algebra, statistics, probability, and calculus.
- Python vs R: Both are good, pick and learn one, and learn the language as a language, not as an ML language. “Data Science from Scratch”, “Machine Learning with R”.
- ML Learning Pipeline: Find data (scrape, APIs, CSVs).
- Machine Learning Mastery website.
- Visualize results. If you can’t communicate your result, you might as well not have done it.
scikit
and algorithm things.- The plan:
- Learn the math (2-3 months).
- Learn the programming language (1 month).
- Machine learning tutorials and test projects (1-2 months).
- Short-term passion projects (1 month).
- You’re going to fail a lot.
- Your models may not be predictive.
- Track your projects on Github and write up your results.