December 2015 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] #sum:, #detectSum:, #sumNumbers:

by Thierry Goubier

Note that the sum with #anyOne is wrong. #(1 2 3 4 5) inject: #(1 2 3 4 5) anyOne into: [ :sum :each | sum + each ] returns 16 instead of 15. Thierry 2015-12-01 10:18 GMT+01:00 Ben Coman <btc(a)openinworld.com>: > On Tue, Dec 1, 2015 at 4:45 PM, Sven Van Caekenberghe <sven(a)stfx.eu> > wrote: > > I am all for a cleanup, the current situation is confusing. > > The basic #sum should be fast AND work for empty collections with 0 as > starting element. > > > I know why the #anyOne is used, and that use case should be preserved, > but it is less common IMHO. > > I'm curious to be learn why? > cheers -ben > > > > >> On 01 Dec 2015, at 09:38, Thierry Goubier <thierry.goubier(a)gmail.com> > wrote: > >> > >> Hi Max, > >> > >> Interesting results... > >> > >> replacing the #yourself in newSum: cuts the runtime by a factor of two. > >> using an [:x | x ] block cost 10% compared to a direct version of sum. > >> > >> So the fastest sum is: > >> > >> sum > >> ^ self > >> inject: > >> (self ifEmpty: [ ^ 0 ] ifNotEmpty: [ self anyOne > ]) > >> into: [ :sum :each | sum + each ] > >> > >> And > >> > >> sum > >> ^ self sum: [:x | x ] > >> > >> is 10% slower. > >> > >> And > >> > >> sum > >> ^ self sum: #yourself > >> > >> is 100% slower (i.e. x2) > >> > >> It may be wise to avoid using #yourself for a block. > >> > >> Thierry > >> > >> > >> 2015-12-01 9:17 GMT+01:00 Max Leske <maxleske(a)gmail.com>: > >> Hi guys, > >> > >> Collection defines #sum:, #detectSum: and #sumNumbers:, all of which > accomplish the same (in principal) but with subtle differences: > >> > >> #sum: > >> - uses #inject:into: with #anyOne as the injected element and will thus > fail for empty collections > >> > >> #detectSum: > >> - uses ânextValue + sumâ instead of âsum + nextValueâ which makes it a > lot slower when dealing with large numbers (primitive fails and number > conversion is necessary) > >> > >> #sumNumbers: > >> - same as #sum but doesnât fail for empty collections. Only good for > numbers though. > >> > >> > >> Benchmarks: > >> > >> [ 100 timesRepeat: [ (1 to: 1000000) sum: #yourself ] ] timeToRun 18062 > >> [ 100 timesRepeat: [ (1 to: 1000000) detectSum: #yourself ] ] timeToRun > 42391 > >> [ 100 timesRepeat: [ (1 to: 1000000) sumNumbers: #yourself ] ] > timeToRun 18096 > >> > >> > >> > >> Can we settle for a single implementation? Such as (modified from > #sumNumbers:): > >> > >> newSum: aBlock > >> ^ self > >> inject: (self > >> ifEmpty: [ 0 ] > >> ifNotEmpty: [ self anyOne ]) > >> into: [ :sum :each | sum + (aBlock value: each) ] > >> > >> This implementation combines the best of the three implementations I > think. Benchmark: > >> > >> [ 100 timesRepeat: [ (1 to: 1000000) newSum: #yourself ] ] timeToRun > 17955 > >> > >> > >> BTW, there is also the message #sum, which suffers from the same > problem as #sum: (the implementation is basically a copy). #sum could be > implemented as > >> > >> sum > >> ^ self sum: [ :x | x ] > >> > >> > >> As for the name for the new method, I suggest #sum:. > >> > >> Cheers, > >> Max > >> > > > > > >

Dec. 1, 2015

Re: [Pharo-dev] #sum:, #detectSum:, #sumNumbers:

by Ben Coman

On Tue, Dec 1, 2015 at 4:45 PM, Sven Van Caekenberghe <sven(a)stfx.eu> wrote: > I am all for a cleanup, the current situation is confusing. > The basic #sum should be fast AND work for empty collections with 0 as starting element. > I know why the #anyOne is used, and that use case should be preserved, but it is less common IMHO. I'm curious to be learn why? cheers -ben > >> On 01 Dec 2015, at 09:38, Thierry Goubier <thierry.goubier(a)gmail.com> wrote: >> >> Hi Max, >> >> Interesting results... >> >> replacing the #yourself in newSum: cuts the runtime by a factor of two. >> using an [:x | x ] block cost 10% compared to a direct version of sum. >> >> So the fastest sum is: >> >> sum >> ^ self >> inject: >> (self ifEmpty: [ ^ 0 ] ifNotEmpty: [ self anyOne ]) >> into: [ :sum :each | sum + each ] >> >> And >> >> sum >> ^ self sum: [:x | x ] >> >> is 10% slower. >> >> And >> >> sum >> ^ self sum: #yourself >> >> is 100% slower (i.e. x2) >> >> It may be wise to avoid using #yourself for a block. >> >> Thierry >> >> >> 2015-12-01 9:17 GMT+01:00 Max Leske <maxleske(a)gmail.com>: >> Hi guys, >> >> Collection defines #sum:, #detectSum: and #sumNumbers:, all of which accomplish the same (in principal) but with subtle differences: >> >> #sum: >> - uses #inject:into: with #anyOne as the injected element and will thus fail for empty collections >> >> #detectSum: >> - uses ânextValue + sumâ instead of âsum + nextValueâ which makes it a lot slower when dealing with large numbers (primitive fails and number conversion is necessary) >> >> #sumNumbers: >> - same as #sum but doesnât fail for empty collections. Only good for numbers though. >> >> >> Benchmarks: >> >> [ 100 timesRepeat: [ (1 to: 1000000) sum: #yourself ] ] timeToRun 18062 >> [ 100 timesRepeat: [ (1 to: 1000000) detectSum: #yourself ] ] timeToRun 42391 >> [ 100 timesRepeat: [ (1 to: 1000000) sumNumbers: #yourself ] ] timeToRun 18096 >> >> >> >> Can we settle for a single implementation? Such as (modified from #sumNumbers:): >> >> newSum: aBlock >> ^ self >> inject: (self >> ifEmpty: [ 0 ] >> ifNotEmpty: [ self anyOne ]) >> into: [ :sum :each | sum + (aBlock value: each) ] >> >> This implementation combines the best of the three implementations I think. Benchmark: >> >> [ 100 timesRepeat: [ (1 to: 1000000) newSum: #yourself ] ] timeToRun 17955 >> >> >> BTW, there is also the message #sum, which suffers from the same problem as #sum: (the implementation is basically a copy). #sum could be implemented as >> >> sum >> ^ self sum: [ :x | x ] >> >> >> As for the name for the new method, I suggest #sum:. >> >> Cheers, >> Max >> > >

Dec. 1, 2015

Re: [Pharo-dev] dual-monitor pharo

by Tudor Girba

Hi, Probably you need to have two different operating system windows. In this case, the work you are probably interested in is OSWindow. For example, when you now switch to a Bloc world, you will get a separate OS window. Cheers, Doru > On Dec 1, 2015, at 8:50 AM, Peter Uhnak <i.uhnak(a)gmail.com> wrote: > > Hi, > > has there been any effort done on running pharo on multiple monitors? > > Is it something that would require deep changes to VM? Can it be done > image-side? > > Thanks, > -- > Peter > -- www.tudorgirba.com "Problem solving should be focused on describing the problem in a way that makes the solution obvious."

Dec. 1, 2015

Re: [Pharo-dev] #sum:, #detectSum:, #sumNumbers:

by Sven Van Caekenberghe

I am all for a cleanup, the current situation is confusing. The basic #sum should be fast AND work for empty collections with 0 as starting element. I know why the #anyOne is used, and that use case should be preserved, but it is less common IMHO. > On 01 Dec 2015, at 09:38, Thierry Goubier <thierry.goubier(a)gmail.com> wrote: > > Hi Max, > > Interesting results... > > replacing the #yourself in newSum: cuts the runtime by a factor of two. > using an [:x | x ] block cost 10% compared to a direct version of sum. > > So the fastest sum is: > > sum > ^ self > inject: > (self ifEmpty: [ ^ 0 ] ifNotEmpty: [ self anyOne ]) > into: [ :sum :each | sum + each ] > > And > > sum > ^ self sum: [:x | x ] > > is 10% slower. > > And > > sum > ^ self sum: #yourself > > is 100% slower (i.e. x2) > > It may be wise to avoid using #yourself for a block. > > Thierry > > > 2015-12-01 9:17 GMT+01:00 Max Leske <maxleske(a)gmail.com>: > Hi guys, > > Collection defines #sum:, #detectSum: and #sumNumbers:, all of which accomplish the same (in principal) but with subtle differences: > > #sum: > - uses #inject:into: with #anyOne as the injected element and will thus fail for empty collections > > #detectSum: > - uses ânextValue + sumâ instead of âsum + nextValueâ which makes it a lot slower when dealing with large numbers (primitive fails and number conversion is necessary) > > #sumNumbers: > - same as #sum but doesnât fail for empty collections. Only good for numbers though. > > > Benchmarks: > > [ 100 timesRepeat: [ (1 to: 1000000) sum: #yourself ] ] timeToRun 18062 > [ 100 timesRepeat: [ (1 to: 1000000) detectSum: #yourself ] ] timeToRun 42391 > [ 100 timesRepeat: [ (1 to: 1000000) sumNumbers: #yourself ] ] timeToRun 18096 > > > > Can we settle for a single implementation? Such as (modified from #sumNumbers:): > > newSum: aBlock > ^ self > inject: (self > ifEmpty: [ 0 ] > ifNotEmpty: [ self anyOne ]) > into: [ :sum :each | sum + (aBlock value: each) ] > > This implementation combines the best of the three implementations I think. Benchmark: > > [ 100 timesRepeat: [ (1 to: 1000000) newSum: #yourself ] ] timeToRun 17955 > > > BTW, there is also the message #sum, which suffers from the same problem as #sum: (the implementation is basically a copy). #sum could be implemented as > > sum > ^ self sum: [ :x | x ] > > > As for the name for the new method, I suggest #sum:. > > Cheers, > Max >

Dec. 1, 2015

Re: [Pharo-dev] #sum:, #detectSum:, #sumNumbers:

by Peter Uhnak

The differences between #sum: and #sumNumbers: are well described in the respective method comments, but I agree that this may be a problem as most people will go for #sum: even for basic numeric collections. (And I personally got bit by this.) So maybe #sum: could adopt #sumNumbers: implementation, and the current #sum: implementation would be moved to #domainSum: (or some other appropriate name...)? As for #detectSum: I think this method has confusing name and is functionally duplicate to #sum:. The swapped arguments there may make sense when you overload #+, however the same thing can be achieved from the provided block ... On 12/01, Max Leske wrote: > Hi guys, > > Collection defines #sum:, #detectSum: and #sumNumbers:, all of which accomplish the same (in principal) but with subtle differences: > > #sum: > - uses #inject:into: with #anyOne as the injected element and will thus fail for empty collections > > #detectSum: > - uses ânextValue + sumâ instead of âsum + nextValueâ which makes it a lot slower when dealing with large numbers (primitive fails and number conversion is necessary) > > #sumNumbers: > - same as #sum but doesnât fail for empty collections. Only good for numbers though. > > > Benchmarks: > > [ 100 timesRepeat: [ (1 to: 1000000) sum: #yourself ] ] timeToRun 18062 > [ 100 timesRepeat: [ (1 to: 1000000) detectSum: #yourself ] ] timeToRun 42391 > [ 100 timesRepeat: [ (1 to: 1000000) sumNumbers: #yourself ] ] timeToRun 18096 > > > > Can we settle for a single implementation? Such as (modified from #sumNumbers:): > > newSum: aBlock > ^ self > inject: (self > ifEmpty: [ 0 ] > ifNotEmpty: [ self anyOne ]) > into: [ :sum :each | sum + (aBlock value: each) ] > > This implementation combines the best of the three implementations I think. Benchmark: > > [ 100 timesRepeat: [ (1 to: 1000000) newSum: #yourself ] ] timeToRun 17955 > > > BTW, there is also the message #sum, which suffers from the same problem as #sum: (the implementation is basically a copy). #sum could be implemented as > > sum > ^ self sum: [ :x | x ] > > > As for the name for the new method, I suggest #sum:. > > Cheers, > Max -- Peter

Dec. 1, 2015

Re: [Pharo-dev] #sum:, #detectSum:, #sumNumbers:

by Thierry Goubier

Hi Max, Interesting results... replacing the #yourself in newSum: cuts the runtime by a factor of two. using an [:x | x ] block cost 10% compared to a direct version of sum. So the fastest sum is: sum ^ self inject: (self ifEmpty: [ ^ 0 ] ifNotEmpty: [ self anyOne ]) into: [ :sum :each | sum + each ] And sum ^ self sum: [:x | x ] is 10% slower. And sum ^ self sum: #yourself is 100% slower (i.e. x2) It may be wise to avoid using #yourself for a block. Thierry 2015-12-01 9:17 GMT+01:00 Max Leske <maxleske(a)gmail.com>: > Hi guys, > > Collection defines #sum:, #detectSum: and #sumNumbers:, all of which > accomplish the same (in principal) but with subtle differences: > > #sum: > - uses #inject:into: with #anyOne as the injected element and will thus > fail for empty collections > > #detectSum: > - uses ânextValue + sumâ instead of âsum + nextValueâ which makes it a lot > slower when dealing with large numbers (primitive fails and number > conversion is necessary) > > #sumNumbers: > - same as #sum but doesnât fail for empty collections. Only good for > numbers though. > > > Benchmarks: > > [ 100 timesRepeat: [ (1 to: 1000000) sum: #yourself ] ] timeToRun 18062 > [ 100 timesRepeat: [ (1 to: 1000000) detectSum: #yourself ] ] timeToRun > 42391 > [ 100 timesRepeat: [ (1 to: 1000000) sumNumbers: #yourself ] ] timeToRun > 18096 > > > > Can we settle for a single implementation? Such as (modified from > #sumNumbers:): > > newSum: aBlock > ^ self > inject: (self > ifEmpty: [ 0 ] > ifNotEmpty: [ self anyOne ]) > into: [ :sum :each | sum + (aBlock value: each) ] > > This implementation combines the best of the three implementations I > think. Benchmark: > > [ 100 timesRepeat: [ (1 to: 1000000) newSum: #yourself ] ] timeToRun 17955 > > > BTW, there is also the message #sum, which suffers from the same problem > as #sum: (the implementation is basically a copy). #sum could be > implemented as > > sum > ^ self sum: [ :x | x ] > > > As for the name for the new method, I suggest #sum:. > > Cheers, > Max >

Dec. 1, 2015

#sum:, #detectSum:, #sumNumbers:

by Max Leske

Hi guys, Collection defines #sum:, #detectSum: and #sumNumbers:, all of which accomplish the same (in principal) but with subtle differences: #sum: - uses #inject:into: with #anyOne as the injected element and will thus fail for empty collections #detectSum: - uses ânextValue + sumâ instead of âsum + nextValueâ which makes it a lot slower when dealing with large numbers (primitive fails and number conversion is necessary) #sumNumbers: - same as #sum but doesnât fail for empty collections. Only good for numbers though. Benchmarks: [ 100 timesRepeat: [ (1 to: 1000000) sum: #yourself ] ] timeToRun 18062 [ 100 timesRepeat: [ (1 to: 1000000) detectSum: #yourself ] ] timeToRun 42391 [ 100 timesRepeat: [ (1 to: 1000000) sumNumbers: #yourself ] ] timeToRun 18096 Can we settle for a single implementation? Such as (modified from #sumNumbers:): newSum: aBlock ^ self inject: (self ifEmpty: [ 0 ] ifNotEmpty: [ self anyOne ]) into: [ :sum :each | sum + (aBlock value: each) ] This implementation combines the best of the three implementations I think. Benchmark: [ 100 timesRepeat: [ (1 to: 1000000) newSum: #yourself ] ] timeToRun 17955 BTW, there is also the message #sum, which suffers from the same problem as #sum: (the implementation is basically a copy). #sum could be implemented as sum ^ self sum: [ :x | x ] As for the name for the new method, I suggest #sum:. Cheers, Max

Dec. 1, 2015

dual-monitor pharo

by Peter Uhnak

Hi, has there been any effort done on running pharo on multiple monitors? Is it something that would require deep changes to VM? Can it be done image-side? Thanks, -- Peter

Dec. 1, 2015

Re: [Pharo-dev] Machine Learning algorithms

by HernÃ¡n Morales Durand

The package NaturalSmalltalk includes some supervised classifiers that you can try: NSBinaryNaiveBayesDocumentClassifier exampleBotchanOrShinran. NSNaiveBayesDocumentClassifier exampleBotchanOrShinran. NSTfidfDocumentClassifier exampleBotchanOrShinran. http://smalltalkhub.com/#!/~hernan/NaturalSmalltalk Cheers, HernÃ¡n 2015-11-30 20:08 GMT-03:00 KlÃ©risson PaixÃ£o <klerissonpaixao(a)gmail.com>: > Greetings! > > We're thinking to power up Pharo with some machine learning algorithms > (Naive Bayes, Random Forest and so on). I'm wondering if there is anything > done already? > > Thanks, > KlÃ©risson >

Dec. 1, 2015

Re: [Pharo-dev] Machine Learning algorithms

by serge.stinckwich＠gmail.com

Great initiative Klerisson ! This is better to talk on the pharo-users I guess. There is a project called SciSmalltalk where a lot of mathematics libraries are already available that might be useful for ML algorithms. Please join us on https://github.com/SergeStinckwich/SciSmalltalk Mailing list here: https://groups.google.com/forum/m/#!forum/scismalltalk Regards Sent from my iPhone > On 1 dÃ©c. 2015, at 00:08, KlÃ©risson PaixÃ£o <klerissonpaixao(a)gmail.com> wrote: > > Greetings! > > We're thinking to power up Pharo with some machine learning algorithms (Naive Bayes, Random Forest and so on). I'm wondering if there is anything done already? > > Thanks, > KlÃ©risson

Dec. 1, 2015