Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Total posts in this thread: 8
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 3386 times and has 7 replies Next Thread
QuantumEthos
Senior Cruncher
Joined: Jul 2, 2011
Post Count: 336
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Program Efficiency : SVM - MACHINE LEARNING : RESEARCH AND PROGRAM : HPC

benchmark CPU 0: fp 2688000000.000000 int 9137190918.864225 intloops 158032000.000000 inttime 9.843750 : CPU 40: fp 2926962457.337884 int 0.000000 intloops 0.000000 inttime 0.000000 SVM ML Spec : Green world Support vector machine, Machine learning and data min

https://www.analyticsvidhya.com/blog/2017/09/...tor-machine-example-code/
https://scikit-learn.org/stable/modules/svm.html
https://towardsdatascience.com/https-medium-c...ushikesh-svm-f4b42800e989
https://towardsdatascience.com/support-vector...a444fca47?gi=51274a92cf9b
http://web.mit.edu/6.034/wwwbob/svm-notes-long-08.pdf

Compression ZLib,Gzip,LZ4a & JS Libraries : https://is.gd/ProcessorLasso

QE https://science.n-helix.com
[May 18, 2020 11:54:39 AM]   Link   Report threatening or abusive post: please login first  Go to top 
QuantumEthos
Senior Cruncher
Joined: Jul 2, 2011
Post Count: 336
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Program Efficiency : SVM - MACHINE LEARNING : RESEARCH AND PROGRAM : HPC

https://science.n-helix.com
# https://docs.microsoft.com/en-us/machine-lear...hat-is-microsoft-r-client
# https://docs.microsoft.com/en-us/machine-lear...oftml/microsoftml-package
# https://docs.microsoft.com/en-us/machine-lear...install-pretrained-models
# https://github.com/iterative/dvc

Revolutions in vector: SVM Machine learning optimised & dynamic point/pointer cached ray tracing
Machine Learning Probability Vector Ray Tracing

ML-PVR-T : Wonderful! ML Probability Vector Ray Tracing
Dynamic Many-Light Sampling for Real-Time Ray Tracing

ReSTIR.pdf (47.85 MB)
https://mirrorace.com/m/X7ia
021-026.pdf (20.02 MB)
https://mirrorace.com/m/X7ib

Machine learning,
The Advanced SVM feature Set & Development

CPU lead Advanced SVM/ML potential
GPU refinement & memory Expansion/Expression/Development

SVM/ML Logic for:
Shaders,
Tessellation,
Compression,
PML Vector Ray-Tracing

Sharpening Image Enhancement:

(S²ecRETA²i)(tm)
Reactive Image Enhancement : ML VSR : Super Sampling Resolution Enhancement with Tessellated Precision & Anti-Aliasing Ai (S²ecRETA²i) + (SSAA)
Color Dynamic Range Quantification, Mesh Tessellation, Smoothing & Interpolation
Finally MIP-MAP optimised sampling with size/distance, dynamic cache compression.

Machine learning,
The Advanced SVM feature Set & New developments..TPU <> GDev,AMD

Extended support for ML means dynamic INT4/8/16/Float types and dot product instructions execution.
GPU/CPU/Feature-set/SVM

Dual compute unit exposure of additional mixed-precision dot-product modes in the ALUs,
Primarily for accelerating machine learning inference,
A mixed-precision FMA dot2 will compute two half-precision multiplications and then add the results to a single-precision accumulator. For even greater throughput,

Some ALUs will support 8-bit integer dot4 operations and 4-bit dot8 operations,
All of which use 32-bit accumulators to avoid any overflows."

Core-ML runs on all 3 hardware parts: CPU, GPU, Neural Engine ASIC;SVM.
The developer doesn’t specify; The software middle-ware chooses which part to run ML models,

Core strategic advice & adaptable SVM CPU <> GPU

(c)QE
[Jun 11, 2020 4:57:34 PM]   Link   Report threatening or abusive post: please login first  Go to top 
hchc
Veteran Cruncher
USA
Joined: Aug 15, 2006
Post Count: 746
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Program Efficiency : SVM - MACHINE LEARNING : RESEARCH AND PROGRAM : HPC

Dude, can you please stop spamming the forums? I try to make a best effort to understand and consider myself fairly educated in both technology and basic science, but I can't make heads or tails out of most of your threads.

At the very least, in plain English, introduce the topic you are trying to discuss before jumping into the details.

Not sure what this content has to do with Suggestions or Feedback, honestly.
----------------------------------------
  • i3-8100 (Coffee Lake, 4C/4T) @ 3.6 GHz
  • i5-4590 (Haswell, 4C/4T) @ 3.3 GHz
  • E5800 (Wolfdale, 2C/2T) @ 3.2 GHz

----------------------------------------
[Edit 1 times, last edit by hchc at Jun 23, 2020 6:37:20 AM]
[Jun 14, 2020 5:18:29 AM]   Link   Report threatening or abusive post: please login first  Go to top 
QuantumEthos
Senior Cruncher
Joined: Jul 2, 2011
Post Count: 336
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Program Efficiency : SVM - MACHINE LEARNING : RESEARCH AND PROGRAM : HPC

I am sick and tired of developing technology for GPL projects that cannot understand that the Suggestions / Feedback forum is for the DATA SCIENTISTS from IBM,

If you cannot code or understand CPU development then please stop trying to understand people who have more than 7 years of hardware & software development.

TIME + Development IS MONEY

QE
[Jun 14, 2020 5:48:03 AM]   Link   Report threatening or abusive post: please login first  Go to top 
hchc
Veteran Cruncher
USA
Joined: Aug 15, 2006
Post Count: 746
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Program Efficiency : SVM - MACHINE LEARNING : RESEARCH AND PROGRAM : HPC

Ok, I think I'm leaning away from robot.

QuantumEthos said:
I am sick and tired of developing technology for GPL projects that cannot understand that the Suggestions / Feedback forum is for the DATA SCIENTISTS from IBM,

I was under the impression this forum was specifically for the WCG user community to provide suggestions or feedback directly to the small but mighty WCG team. tongue

You have many threads (dozens?) in the Hardware Forum that have similar content. Some are longer or shorter than others, but most have a link to your blog. Many of the threads are very unreadable and difficult to follow and just seems like a bunch of spam links and copy-pasta. It's precisely why they often have 0 replies -- people think it's spam or unreadable.

QuantumEthos said:
If you cannot code or understand CPU development then please stop trying to understand people who have more than 7 years of hardware & software development.

I'm trying to respond as civilly as possible while mantaining my privacy and humility, but I learned programming as a kid in the early 90s. That's a bit more than seven years, but I don't have my abacus. I'm still learning as I go of course. Also as a teenager I was honored to work my tail off and win two national championships in computing and technology events. Lately I've worked in IT for quite a few years, so my development skills are limited to writing scripts and automation -- nothing "heavy" like a full-time software developer. I have both an engineering coursework background as well as some coursework in the basic natural sciences, and a fairly robust computing background. I haven't, however, dabbled in "hardware development" or "CPU development" as you also put it, so hopefully I can learn from you.

That said, I wouldn't necessarily assume that you are the smartest or most experienced in the room. It's much better to be humble and courteous to everybody.

I wasn't trying to attack you with that meme or my post. It's just you seem to keep making threads along the same vein, and they're very spammy and extremely difficult to understand. I replied to one of your threads in the Hardware forum in 2018 perhaps, and you simply linked to Microsoft web pages on existing security vulnerabilities in Windows and somehow (in your mind?) you made it seem like there was something actionable for either the user community or the WCG team in there. It was really just spam and nothing earth-shattering.

QuantumEthos said:
TIME + Development IS MONEY

QE


Ok man, carry on. I'm not alone in thinking your threads are super hard to follow (bordering on unintelligible), and perhaps some self-awareness would be good for you that maybe the way you are communicating your ideas isn't effectively reaching your intended audience (whether that's us users, the WCG team, or "DATA SCIENTISTS from IBM.")

Cheers dude! coffee whistling
----------------------------------------
  • i3-8100 (Coffee Lake, 4C/4T) @ 3.6 GHz
  • i5-4590 (Haswell, 4C/4T) @ 3.3 GHz
  • E5800 (Wolfdale, 2C/2T) @ 3.2 GHz

----------------------------------------
[Edit 2 times, last edit by hchc at Jun 14, 2020 6:41:14 AM]
[Jun 14, 2020 6:34:59 AM]   Link   Report threatening or abusive post: please login first  Go to top 
QuantumEthos
Senior Cruncher
Joined: Jul 2, 2011
Post Count: 336
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Program Efficiency : SVM - MACHINE LEARNING : RESEARCH AND PROGRAM : HPC

<3 ôo Ah yes the problem is Links have pages of documents..

For example R Python is at least 40 pages of document and downloads.

QE
[Jun 14, 2020 7:07:59 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Program Efficiency : SVM - MACHINE LEARNING : RESEARCH AND PROGRAM : HPC

I am sick and tired of developing technology for GPL projects that cannot understand that the Suggestions / Feedback forum is for the DATA SCIENTISTS from IBM


First, I would submit that the Suggestions/Feedback forum is for the WCG Techs, not the DATA SCIENTISTS at IBM. I would suggest that the forum was created for members to make suggestions and give feedback related to WCG.

Second, respectfully, I don't think IBM needs your assistance in developing anything. Their Distinguished Engineers hold many patents. I have an idea, if you want to help IBM and feel you have knowledge that they don't already have, apply for a job. Your posts provide nothing useful for members or the techs. They are too dis-jointed, choppy, and unintelligible. Additionally, they don't contain any of your thoughts, they only contain links to writings of others. Essentially, you are a research assistant.
----------------------------------------
[Edit 1 times, last edit by Former Member at Jun 14, 2020 3:26:27 PM]
[Jun 14, 2020 3:25:46 PM]   Link   Report threatening or abusive post: please login first  Go to top 
QuantumEthos
Senior Cruncher
Joined: Jul 2, 2011
Post Count: 336
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Program Efficiency : SVM - MACHINE LEARNING : RESEARCH AND PROGRAM : HPC

I will have you know that Vulkan is the work of AMD and myself as a source on a base of AMD's metal configuration...

Without AMD Vulkan would not exist, As to myself you are to read the following and note the RS in the ends of windows builds.. you are also to note VRS and other RS'y things and take the hint.

some of my personal contributions :

read all

https://science.n-helix.com/2019/06/kernel.html

https://science.n-helix.com/2019/06/vulkan-stack.html

https://science.n-helix.com/2018/01/integer-floats-with-remainder-theory.html

https://science.n-helix.com/2016/04/3d-desktop-virtualization.html

https://science.n-helix.com/2017/02/open-gaming.html

https://science.n-helix.com/2019/05/compiler-optimisation.html

https://science.n-helix.com/2020/01/float-hlsl-spir-v-compiler-role.html

Compiler books & reading : https://science.n-helix.com/2017/04/boinc.html
[Jun 14, 2020 11:42:07 PM]   Link   Report threatening or abusive post: please login first  Go to top 
[ Jump to Last Post ]
Post new Thread