Hi all!
Some weeks ago I found the article: Comparing K-Means and Others Algorithms for Data Clustering (https://www.codeproject.com/Articles/5375470/Comparing-K-Means-and-Others-Algorithms-for-Data-C) by Nicolás Descartes. With C# source code.
Look very interesting to translate that to Assembly. Still is a work in progress but look well.
There is some kind of abuse of Collections, because is very easy to write that in C#. I removed most obvious exagerations (perhaps they try to make more clear the algorithm, not sure).
- K-Means strategy need initial randomness, then probably you have to run several times before to find the better solution.
- Hierarchical strategy need to define termination, and in these cases are number of clusters.
This 2 don't need collections indeed, but I follow the author here. I use a K-Means method I wrote in PowerBasic for DOS three weeks ago :biggrin:
- Density-based spatial clustering of applications with noise really benefit from Collections, and also need a couple of Sorted Vectors.
Using this vectors happen that .VecForEach/.VecNext have some problems if you use more than one kind of vector.
Anyway is a good challenge, and still I have to collect some leaks :biggrin:
Any sugestion or improvement is welcome!
Regards, HSE
Source Code and 64bit Binary updated 9 March 2024 in GitHub (https://github.com/ASMHSE/Clusters-in-Assembly/tree/main)
.
Hi HSE
Very interesting and cool project, I'll read the CP article first to get a deeper understanding. :thumbsup:
QuoteUsing this vectors happen that .VecForEach/.VecNext have some problems if you use more than one kind of vector.
After reading the code, I think I understand the problem. Since the vectors are of different sizes, the macros don't take this into account, with disastrous results. The good news is that we can do better. :cool:
Biterider
Quote from: HSE on January 30, 2024, 07:36:03 AMAny sugestion or improvement is welcome!
Use the F1 key :biggrin:
Hi Biterider!
Quote from: Biterider on January 30, 2024, 08:33:36 AMVery interesting and cool project, I'll read the CP article first to get a deeper understanding. :thumbsup:
Thanks :thumbsup:
Quote from: Biterider on January 30, 2024, 08:33:36 AMThe good news is that we can do better. :cool:
Yes. Just my first idea fail :biggrin: :biggrin: , and is not necessary right now.
In comparison, .ColForEach/.ColNext was critical, and work
"impecable".
HSE
Quote from: jj2007 on January 30, 2024, 10:37:43 AMUse the F1 key :biggrin:
Good idea :thumbsup:
It's not in my rudimentary template (nor is the help :biggrin: ).
Thanks
Hi HSE
I'm trying to compile the "Clusters" application, but I get an error that I can't fix:
forced error: [back end: x86-32/64,FPU][#CG6] function is undefined or not supported by current back end: ceil
I downloaded the latest version of SmplMath from GitHub, but that didn't solve the problem. Also, there seems to be a missing file called IfElseM.inc, which I had to borrow from another previous installation.
Without being able to compile the above application, I tried to solve the TVector problem, assuming that the problem was due to different element sizes, and prepared a modification that might solve it.
Would you give them a try?
Regards, Biterider
Hi Biterider!
Quote from: Biterider on January 31, 2024, 05:29:14 AMAlso, there seems to be a missing file called IfElseM.inc, which I had to borrow from another previous installation.
:biggrin: :biggrin: Thanks
In math.inc you can change "IfElseM.inc" by "FlowControl.inc", but I think nothing of that is used.
Quote from: Biterider on January 31, 2024, 05:29:14 AMforced error: [back end: x86-32/64,FPU][#CG6] function is undefined or not supported by current back end: ceil
In math_functions.inc you can add:
fslv_fnc_ceil macro
IFE fslv_volatile_gprs AND FSVGPR_EAX
T_EXPR(<push eax>,<mov [rsp+8],rax>)
ENDIF
fstcw WORD ptr T_EXPR([esp-2],[rsp])
movzx eax,WORD ptr T_EXPR([esp-2],[rsp])
or eax,0800h
mov T_EXPR([esp-4],[rsp+2]),ax
fldcw T_EXPR([esp-4],[rsp+2])
frndint
fldcw WORD ptr T_EXPR([esp-2],[rsp])
IFE fslv_volatile_gprs AND FSVGPR_EAX
T_EXPR(<pop eax>,<mov rax,[rsp+8]>)
ENDIF
endm
default_fnc_dscptr2 <ceil>,nArgs=1,fpu=-1,x64=-1
Currently I use the Complete SmplMath package, with DoubleDouble precision and complex numbers, but package size and preprocess are bigger. Perhaps I have to post a Full version, that nobody will use :biggrin: :biggrin:
Previously, .VecForEach don't know about first kind of vector, now about last vector.
In .VecNext apparently must be
inc dword ptr @CatStr(<??VecForEach_Index_>, %??VecForEach_ID)Regards, HSE
Thanks HSE
Now I can compile the application. :thumbsup:
Could you show me the places where you wanted to use the TVector macros?
Biterider
Quote from: HSE on January 31, 2024, 07:12:23 AMPerhaps I have to post a Full version, that nobody will use :biggrin: :biggrin:
At least you can count with one :thumbsup:
Biterider
Quote from: Biterider on January 31, 2024, 08:04:51 AMCould you show me the places where you wanted to use the TVector macros?
Was for testing porpouses, because for debugging I used the classic Randy loop.
StrategyForClusters.inc line 597 and line 629. For this sorted vector is TVectorName
S
Quote from: jj2007 on January 30, 2024, 10:37:43 AMUse the F1 key :biggrin:
Updated in first post with F1 and Alt+F1 keys
The help is a PDF from Word, very big. I will have to build fron T
EX :biggrin:
Quote from: HSE on January 31, 2024, 08:13:35 AMStrategyForClusters.inc line 597 and line 629.
Looking at these lines, I think I have it now.
The best way to provide the missing information is to use something like this (for vectors only)
.VecForEach [ebx]::Real8VectorS
...
.VecNext
If there are no objections, I will have a go at it. :biggrin:
Biterider
Quote from: Biterider on January 31, 2024, 08:49:22 AMIf there are no objections,
Mmm :rolleyes: ... no
:thumbsup:
Quote from: HSE on January 31, 2024, 08:40:44 AMThe help is a PDF from Word, very big
You are right, it's far too big. Add the two attached files to your folder, and when user presses VK_F1, do an
invoke WinExec, chr$("clusters.rtf"), SW_RESTORE :cool:
Side comment:
PDF from Word? Gag, puke, retch, barf. Have you ever looked at one of those abominations? Boy, Micro$oft really screwed up with that conversion! Yikes!
We now return you to your regularly scheduled thread.
Quote from: NoCforMe on January 31, 2024, 11:55:58 AMPDF from Word? Gag, puke, retch, barf
:joking: :rofl: :greenclp:
Pdf is crap, indeed. I managed to convert pdf to rtf - just run the attached exe to see HSE's help file in all its beauty (btw the exe was definitely
not built with MasmBasic, hehe - look at the size) :cool:
From 311,187 to 4,608 bytes - the power of Assembly!
Quote from: jj2007 on January 31, 2024, 12:48:04 PMPdf is crap, indeed.
Just to be clear, PDFs from Word docs are crap. Other PDFs are fine.
I used to own a print shop (ca. 2005). I was so happy when customers would bring in their jobs to print as PDFs; I could just place it in Adobe InDesign, ship it to my platemaker and off I'd go, perfect every time.
Thanks JJ!
Quote from: jj2007 on January 31, 2024, 12:48:04 PMFrom 311,187 to 4,608 bytes - the power of Assembly!
Not so bad, except that I don't like very much RTF files :biggrin:
But if using T
EX still PDF file is big, I could store RTF internally. I made that with a help in TXT format in MathArtH.
With LaTEX resulting PDF file have 54 Kb, a lot more reasonable, and look good :thumbsup:
Updated in first post.
Hi HSE
I have created a test version with your source files. I have changed the interaction macros and added a variable to the object definition that is filled at compile time.
When I run the binary with these changes, it seems to give reasonable results on DC.
Perhaps you can interpret this better than I can. :rolleyes:
Biterider
Hi Biterider!
Quote from: Biterider on February 01, 2024, 07:10:14 AMPerhaps you can interpret this better than I can. :rolleyes:
Look perfect :thumbsup:
Thanks, HSE