|
that seems par for the course with haskell e: i thought java was v bad, but after working with java i'd say it's thoroughly ok HoboMan fucked around with this message at 00:38 on Apr 15, 2016 |
#
?
Apr 15, 2016 00:35
|
|
|
|
| # ? May 24, 2024 22:00 |
|
|
Barnyard Protein posted:that's how I feel all the time when programming racket. it's so easy to slowly build up something that does something useful, then at the end have no clue how it works same with me and embedded junk my experience is in things that talk to databases and web browsers so this OS scheduler junk is magic to me
|
|
#
?
Apr 15, 2016 00:37
|
|
|
Notorious QIG posted:and trainFromCsv() trains the network from a .csv i really love it when the fiddly implementation details of reading a csv (do you use a library for this, or is yet-another-bespoke-csv-like-format?) occupy the same space as my machine-learning neural net stuff. i can think of no reason you'd ever want to separate those two wildly distinct concepts. none at all.
|
|
#
?
Apr 15, 2016 00:42
|
|
|
Jabor posted:i really love it when the fiddly implementation details of reading a csv (do you use a library for this, or is yet-another-bespoke-csv-like-format?) occupy the same space as my machine-learning neural net stuff. i can think of no reason you'd ever want to separate those two wildly distinct concepts. none at all.
|
|
#
?
Apr 15, 2016 01:08
|
|
|
i am doing code things again in free time yosbbs is back and i am going to hack on it rabbitmq and memcached may be involved D:
|
|
#
?
Apr 15, 2016 01:10
|
|
|
I used 0mq once it was nice
|
|
#
?
Apr 15, 2016 01:11
|
|
|
Jabor posted:i really love it when the fiddly implementation details of reading a csv (do you use a library for this, or is yet-another-bespoke-csv-like-format?) occupy the same space as my machine-learning neural net stuff. i can think of no reason you'd ever want to separate those two wildly distinct concepts. none at all. it's straight up an stl-based csv reader and the main training loop is a function called propagate and a function called back propagate it's v easy to extend to new formats
|
|
#
?
Apr 15, 2016 01:22
|
|
|
wish I could back propagate your posts
|
|
#
?
Apr 15, 2016 01:34
|
|
|
Corla Plankun posted:basically everything relating to distributed computation is in java or sometimes a java-based other language, i do not know why they had a lot of work to do for distributed garbage collection (and oh god all these papers are the most terrifying thing in the world, while simultaneously being very boring) is my guess.
|
|
#
?
Apr 15, 2016 01:35
|
|
|
why not erlang
|
|
#
?
Apr 15, 2016 01:36
|
|
|
memcache deez nuts
|
|
#
?
Apr 15, 2016 01:56
|
|
|
penus de milo posted:memcache deez nuts 
|
|
#
?
Apr 15, 2016 02:22
|
|
|
gonadic io posted:What other programming experience do you have? i have a lot of experience with python and ruby and i worked in embedded dev for a few years wriing C before that. and i wrote a java thing once for a computer networking class once to demonstrate some basic-bitch protocol that i cant remember i think i am just going to look around github for something 'neat' in java with not that many locs because i am sure i could catch on fast once i get over my disgust (which probably stems from the fact that the first lang i tried to learn was java in like 1996 and i couldn't figure it out so instead i coded exclusively in mushClient until i finally took a computer programming course in high school)
|
|
#
?
Apr 15, 2016 03:03
|
|
|
Bloody posted:wish I could back propagate your posts gently caress me im so owned
|
|
#
?
Apr 15, 2016 05:03
|
|
|
Notorious QIG posted:again because it's a simple example to demonstrate the general gist of what im talking about you have to use "butt" and "fart" as your variable names or us terrible yospos programmers will think you're posting production code.
|
|
#
?
Apr 15, 2016 07:49
|
|
|
did someone say fart?
|
|
#
?
Apr 15, 2016 08:20
|
|
|
Bloody posted:once my parser-linter works well enough i am going to create verilog++ and here I figured you were doing this to integrate clash with existing Verilog
|
|
#
?
Apr 15, 2016 08:24
|
|
|
So one of our services regularly sends us emails with the body and subject "null". Nobody wants to investigate.
|
|
#
?
Apr 15, 2016 10:12
|
|
|
AWWNAW posted:try saving your changes every once in a while. that happens automatically when you build. vs also has auto recover yeah lol if you aren't subconsciously hitting save after each line of changes
|
|
#
?
Apr 15, 2016 10:50
|
|
|
eschaton posted:and here I figured you were doing this to integrate clash with existing Verilog clash scares me
|
|
#
?
Apr 15, 2016 11:50
|
|
|
Bloody posted:and also i have no idea what i am doing plz don't share the fine print of my email sig
|
|
#
?
Apr 15, 2016 12:29
|
|
|
gonadic io posted:So one of our services regularly sends us emails with the body and subject "null". Nobody wants to investigate. setup a filter in outlook and move on. problem solved
|
|
#
?
Apr 15, 2016 12:57
|
|
|
Wheany posted:you have to use "butt" and "fart" as your variable names or us terrible yospos programmers will think you're posting production code. what if i write enterprise code for the fartbutt industry
|
|
#
?
Apr 15, 2016 13:44
|
|
|
YeOldeButchere posted:setup a filter in outlook and move on. problem solved thus solving the problem once and for all!
|
|
#
?
Apr 15, 2016 14:18
|
|
|
lmao it turns out i am incredibly stupid. all i had to do to get current parser position is add a location <- getPosition line to my do blocks now i have to refactor all of my data types to carry around their location i guess?
|
|
#
?
Apr 15, 2016 14:38
|
|
|
Wheany posted:you have to use "butt" and "fart" as your variable names or us terrible yospos programmers will think you're posting production code. jokes on you guy, i use butt and fart in production
|
|
#
?
Apr 15, 2016 14:42
|
|
|
gonadic io posted:So one of our services regularly sends us emails with the body and subject "null". Nobody wants to investigate. this forever
|
|
#
?
Apr 15, 2016 14:53
|
|
|
i have created a revolting clusterfuck of types and it is a mess
|
|
#
?
Apr 15, 2016 14:54
|
|
|
Notorious QIG posted:what if i write enterprise code for the fartbutt industry are you hiring? i'm willing to relocate
|
|
#
?
Apr 15, 2016 15:26
|
|
|
Bloody posted:i have created a revolting clusterfuck you're posts!!
|
|
#
?
Apr 15, 2016 15:35
|
|
|
Wheany posted:are you hiring? i'm willing to relocate we currently have job offerings on uranus
|
|
#
?
Apr 15, 2016 15:36
|
|
|
HoboMan posted:that seems par for the course with haskell java the language is ok but when added to the tools and frameworks its good.
|
|
#
?
Apr 15, 2016 15:37
|
|
|
Progressive JPEG posted:yeah lol if you aren't subconsciously hitting save after each line of changes control + k, d control + s after typing anything
|
|
#
?
Apr 15, 2016 15:39
|
|
|
Java is okay I've never used Haskell
|
|
#
?
Apr 15, 2016 15:42
|
|
|
Notorious QIG posted:what if i write enterprise code for the fartbutt industry the evil fartbutt industrial complex
|
|
#
?
Apr 15, 2016 15:49
|
|
|
i just started using dropwizard for java and it seems good. about as easy as django or rails except you have to hit auto complete a bunch in your IDE.
|
|
#
?
Apr 15, 2016 16:30
|
|
|
Shaggar posted:control + k, d auto format can happen automagically now (VS2015?) whenever you close a curly or type a semicolon I think. another thing they stole from reshaper
|
|
#
?
Apr 15, 2016 16:34
|
|
|
today i learned that generating reports was so slow because the report function workflow was: 1. query the database for some extra information 2. use that info to insert a new row in a table called "report records" 3. query the database for the actual report info by pulling everything through the "report records" table (no info from this table is actually used in the report) 4. redundantly filter the records received by a bunch a parameters that were already used in the select
|
|
#
?
Apr 15, 2016 16:38
|
|
|
Flat Daddy posted:i just started using dropwizard for java and it seems good. about as easy as django or rails except you have to hit auto complete a bunch in your IDE. use asp.net webapi instead.
|
|
#
?
Apr 15, 2016 16:44
|
|
|
|
| # ? May 24, 2024 22:00 |
|
|
Bloody posted:lmao it turns out i am incredibly stupid. all i had to do to get current parser position is add a location <- getPosition line to my do blocks i haven't paid a lot of attention to what you were doing, but in my toy compiler, during the parsing pass i instead had each data type's parse method return its error information (including location) to the caller so that the errors bubbled up the tree (which also inherently deduplicates redundant errors) and that way the root data type has a complete listing of the errors and their locations and you can just print and quit right there. it sounds like you're thinking of storing your errors in place and then re traversing everything to actually output them, which is gonna add a constant factor of 2 to the work you're doing.
|
|
#
?
Apr 15, 2016 17:16
|
|
