I wish this was exaggerated, but it isn’t at all. Every time I try to learn Haskell, I end up in some tutorial: “You know how you sometimes need to represent eigenvectors in an n-dimensional plane with isotonically theoretical pulsarfunctions? Haskell types make that easy!”
This is the kind of pointless comment I see in my codebase all the time. Best I can tell, a couple of my coworkers like to plan out their code using comments, then backfill in the actual executable code. That’s fine, but they leave the comments in when they add no value.
` public static LocalDate parseEndDateFromString(String dateString) {
<span style="color:#323232;"> try {
</span><span style="color:#323232;">
</span><span style="color:#323232;"> String[] split = dateString.split("-");
</span><span style="color:#323232;">
</span><span style="color:#323232;"> //In order to get the last day of the desired month, we go to the first day of the next month, account for rollover, then subtract one day
</span><span style="color:#323232;">
</span><span style="color:#323232;"> int month = Integer.parseInt(split[0]) == 12 ? 1 : Integer.parseInt(split[0]) + 1;
</span><span style="color:#323232;">
</span><span style="color:#323232;"> return LocalDate.of(Integer.parseInt(split[1]), month, 1).minusDays(1);
</span><span style="color:#323232;">
</span><span style="color:#323232;"> } catch (Exception e) {
</span><span style="color:#323232;">
</span><span style="color:#323232;"> throw new RuntimeException("Invalid date format - must be MM-YYYY");
</span><span style="color:#323232;">
</span><span style="color:#323232;"> }
</span><span style="color:#323232;">
</span><span style="color:#323232;">}`
</span>
Stuff like this, otoh, is where comments are useful. The required format is obvious from the error message, the param and return from the method signature, the only part that requires a comment is the fiddly logic of accounting for the edge case where month == 12 and the rationale behind how we determine the last day of the month. As a rule, comments are for why something is being done, if it’s not obvious, and for magic numbers. Code should tell you what code does.
edit: can anyone spot the bug that I introduced with that parseEndDateFromString() method?
You get a horse and arrive at the castle within seconds but the horse is too old and doesn’t work with the castle.
You remove the horse, destructure the castle and rescue the princess within seconds, but now you have no horse.
While you’re finding a compatible horse and thinking whether you should write your own horse, Bowser recaptures the princess and moves her to another castle.
I had a friend at university who got a job fixing cobol stuff before Y2K. The bank paid him extremely well, housed him in a luxury apartment during the job, and, as he had no driving licence, dropped in a car with free driver for him.
At the simplest, it takes in a vector of floating-point numbers, multiplies them with other similar vectors (the “weights”), sums each one, applies a RELU* the the result, and then uses those values as a vector for another layer with it’s own weights (or gives output). The magic is in the weights.
This operation is a simple matrix-by-vector product followed by pairwise RELU, if you know what that means.
Where modelWeights is [[[Float]]], and so layer has type [Float] -> [[Float]] -> [Float].
RELU: if i>0 then i else 0. It could also be another nonlinear function, but RELU is obviously fast and works about as well as anything else. There’s interesting theoretical work on certain really weird functions, though.
Less simple, it might have a set pattern of zero weights which can be ignored, allowing fast implementation with a bunch of smaller vectors, or have pairwise multiplication steps, like in the Transformer. Aaand that’s about it, all the rest is stuff that was figured out by trail and error like encoding, and the math behind how to train the weights. Now you know.
Assuming you use hex values for 32-bit weights, you could write a line with 4 no problem:
That’s cool, though honestly I haven’t fully understood, but that’s probably because I don’t know Haskell, that line looked like complete gibberish to me lol. At least I think I got the gist of things on a high level, I’m always curious to understand but never dare to dive deep (holds self from making deep learning joke). Much appriciated btw!
Yeah, maybe somebody can translate for you. I considered using something else, but it was already long and I didn’t feel like writing out multiple loops.
No worries. It’s neat how much such a comparatively simple concept can do, with enough data to work from. Circa-2010 I thought it would never work, lol.
programmer_humor
Top
This magazine is from a federated server and may be incomplete. Browse more on the original instance.