« first day (1599 days earlier)      last day (172 days later) » 

RGS
10:24 AM
I can't wait to see all of your solutions to P27. This problem has given me such a headache :P
Announcement: there's a bioinformatics programming competition starting today and we can use whatever language(s) we like: stepik.org/course/91751/syllabus
 
10:47 AM
cool!
looks similar to rosalind or something where you just upload the answer, the code doesn't matter
 
RGS
Yup, that's because Rosalind is either sponsoring the comp or outright giving them the problems.
I read the "Rosalind" name somewhere.
 
ah lol
well I'll definitely give it a go
 
RGS
me too
 
the prizes are strange, I'm not sure I'd want my genome sequenced, sounds a bit dystopian
 
RGS
If you win, I don't think they will come after you to force you into giving them a sample of your DNA
On the other hand, if you wanted that, what would you do..? Spit into a cup and mail it to them? :P
 
10:52 AM
yeah probably that would be weird
 
 
1 hour later…
11:54 AM
we know you have the DNA, rak1507
 
what if a robot wins
 
12:07 PM
these challenges would be much easier with scripting and being able to pipe to stdin
 
 
2 hours later…
RGS
1:52 PM
Problem 1 is worded in such a stupid way, IMO
 
where
 
yea it's a little bit confusing
easy problem though
 
RGS
@Razetime stepik.org/course/91751/syllabus bioinformatics contest
 
@rak1507 Why, you got something to hide?
 
RGS
Woops, ⎕IO←0 just bit me
 
2:06 PM
@user for legal reasons I decline to comment
 
RGS
It's hilarious that problem 2 looks so similar to the problem discussed here yesterday.
 
yeah lol
 
RGS
2:22 PM
@rak1507 sorry to bother, but can you confirm with me that the sample output of that problem is wrong?
I think the 5th row of the sample output is wrong.
 
mj​+ak​>0
 
RGS
@rak1507 ahhhh
ty
 
I did a crap bruteforce solution to pass a few tests but it leaves me 1 mark off qualifying for the next stage lol, obviously deliberate
have to actually do it properly
 
RGS
@rak1507 oh, so bruteforce ain't gonna cut it? :(
 
nope
 
RGS
2:25 PM
Then I guess I'll have to read yours and dzaima's solutions to yesterday's problem.
 
haha
damn, still not efficient enough for test case 3, despite doing 2 a lot faster
 
2:45 PM
@RGS yeah idk what it even means
funny coincidence
 
RGS
@Razetime they want you to count unique columns
 
oh
i was (kinda) getting at that
@dzaima does dzaima/APL have file IO
 
@Razetime you can read lines of a file with ⎕lns'name' and write a string (not an array of lines for whatever reason) with 'name'⎕lns'string'
 
cool
so I'll have to join with ⎕UCS 10 everywhere i guess
 
yeah. No clue why there isn't a whole file read quad.
 
2:54 PM
is name relative to program directory or absolute?
 
relative to the directory where dzaima/APL was invoked (it can of course be absolute too, if you want that)
 
relative is all i need
 
yay I passed test #3
my code is disgusting
 
I just want an apl which i can use with a normal editor without cardiac arresting
 
one of the lines is 170 chars long
 
3:01 PM
wow
 
tbf a lot of it is boilerplate to do parsing + stuff
 
hm there must be some golfing language with file i/o and decent docs
 
yes! passed test #4 as well
unfortunately not part 5 bc they're all 10k :( need something much better than what I have atm
 
3:51 PM
hm if you passed all those you've already qualified right
 
yep
 
I'm gonna try finishing all of them for funsies
 
rip, I thought I had a nice solution for p3 but it is slow for d=100 and q=100 and the first test case is d=q=5000 lol
 
haven't even read that yet
 
it looks more confusing than it is
pen and paper helps
 
3:59 PM
thankfully i have a lot of that nearby due to exams
 
 
3 hours later…
7:04 PM
and in the end it mentions an expression
R←((V⍳V)=⍳⍴V)/V
I have created a vector in tryapl and tried the expression with it but I don't understand what it is doing
 
unique elements (nowadays you'd use ∪)
 
what I understand so far is that ⍳⍴V gives me a vector of indices in related to the size of the vector
aaah!
thank you @rak1507
 
V⍳V is the index of the first occurrence of each thing, ⍳⍴V are the indices, where the two are equal are where the item occurs for the first time
and then / to filter
 
beautiful
thanks a lot for the clarification
 
np
 
 
1 hour later…
8:34 PM
yay, I solved question 5, not fast or pretty but it works
 

« first day (1599 days earlier)      last day (172 days later) »