Re: Optimization can be tricky

Nicethe proof is in the pudding. Thats why the say a good programmer spends most of their time staring at the ceiling ... thinking. Who they are ...I do not know. On Tue, Jun 19, 2018 at 12:47 AM, Curry Kenworthy via use-livecode < use-livecode@lists.runrev.com> wrote: > > Geoff wrote: >

Re: Optimization can be tricky

Geoff wrote: > This is certainly not identical to the original method, but it's > close enough, and runs in a small fraction of a second. Good approach! Optimized code rules. Best wishes, Curry K. ___ use-livecode mailing list use-livecode@lists.

Re: Optimization can be tricky

Thanks for all the suggestions -- I tried multiple ideas without much improvement, then decided to rethink the problem. Roughly: I have a bunch of source data that I categorized by two numbers, the first from 1-N where N might be anywhere from 100-5,000, and the second from 1-100. For any given fi

Re: Optimization can be tricky

Thanks for the tip Ralphlove the sound of that filer function. On Tue, Jun 12, 2018 at 7:00 PM, Curry Kenworthy via use-livecode < use-livecode@lists.runrev.com> wrote: > > Optimizing scripts in LC is not the same as running reports in other > software suites. If you only need 10 results, you

Re: Optimization can be tricky

Optimizing scripts in LC is not the same as running reports in other software suites. If you only need 10 results, you probably don't want to handle all items twice. I hate to loop through all items even once. But if I do, I may be done! I'm not making a full report to print out; I'm just g

RE: Optimization can be tricky

ssage- From: use-livecode [mailto:use-livecode-boun...@lists.runrev.com] On Behalf Of hh via use-livecode Sent: Tuesday, June 12, 2018 4:39 PM To: use-livecode@lists.runrev.com Cc: hh Subject: Re: Optimization can be tricky The scenario Geoff described is roughly to get the top ten (a handful) of

Re: Optimization can be tricky

Then there's something about Geoff's problem (or problem statement) that I don't understand. What on earth is in those records for sorting a mere 2000 of them to be noticable? (I had thought he had to be talking about magnitudes more lines than that). LC (9.0) sorts 20,000 lines of >1000 cha

Re: Optimization can be tricky

The scenario Geoff described is roughly to get the top ten (a handful) of 2000 records comparing a certain numeric value of each. To get that unknown value of each one has to go once through all the records *and then sort* for that ranking. (LiveCode is very fast with a simple numeric sort!) Any ot

Re: Optimization can be tricky

Put yourself in the computer's shoes, and also clarify what you need to accomplish. You are asking it to sort the entire list of 2000 records, but (if I understand) you only want a handful of those. And it has already gone through all the records once before the sort. If you asked a human t

Re: Optimization can be tricky

I don't know if these are good ideas or BAD ideas and without some suitable data, not sure if I could find out - so I'll throw the ideas over and let you decide if they are worth trying :-) 1. Two ideas combined  - avoid scanning for the "item 1" of each line  - use the (hopefully optimi

Re: Optimization can be tricky

Sorry, I forgot to mention Jerry who already proposed to pull out computations from the inner loop and also to use variables in the inner loop. My experience says to avoid getting items as often as possible. And to use the random() in the inner repeat loop instead of in the final sort may be worth

Re: Optimization can be tricky

You could try the following: repeat for each key T in interestArray[uID] put item 1 of interestArray[uID][T] into i1 put item 2 of interestArray[uID][T] into i2 repeat for each line S in storyArray[T] put userSeenArray[uID][item 1 of S] into s1 put abs(item 2 of S - i1) into s2 i

Re: Optimization can be tricky

Hi Geoff, One thing to try in your original code, which should be significantly faster if the array is big, is using > repeat for each key T in interestArray[uID] instead of > repeat for each line T in the keys of interestArray[uID] The latter has to allocate memory for a string containing all

RE: Optimization can be tricky

I know that this might be a different use-case but I have a CA Lottery - Fantasy five lottery parser that collects lottery data and loads it into a matrix and provides the total number of hits for each number. This also has 4 optimization buttons to the show differences. This aupplication was o

Re: Optimization can be tricky

Do you need the entire list sorted randomly or are you just needing to select N random entries from the list (weighted)? How does the speed change if you do a simple numeric sort? You could use a function on a random number to weight things - would need to work out the right math to get what yo

Re: Optimization can be tricky

At first glance, it looks like you might save some time by grabbing interestArray[uID][T] as soon as you have T, and then use what you grabbed in the 3 later places instead of re-figuring interestArray[uID][T] each time. As in: repeat for each line T in the keys of interestArray[uID] put in

Re: Optimization can be tricky

please do goeff this subject is very interesting to me. i have some problems where i need to optimize in a similar way. which kind of repeat look have u found works fastest? have u tried ? repeat for each key this_key in array? is that slower? i love saving milliseconds. :) makes a big diff