Thursday, August 25, 2016

Digest for comp.lang.c++@googlegroups.com - 15 updates in 6 topics

Ramine <ramine@1.1>: Aug 25 07:15PM -0400

Hello.....
 
Hope you have read my previous post about artificial intelligence...
 
Why am i posting in this newsgroup about artificial intelligence?
because i think hardware and software enginneers must know the basics of
artificial intelligence.. this is why as a software programmer
i have learned more about Deep Learning in artificial intelligence..
 
The basic of Deep Learning in artificial intelligence is that you have
to know about the the chain rule in mathematics and how to do partial
derivatives and how to calculate a stochastic gradient descent .. in
Perceptron and Backpropagation neural networks you have to minimize the
error of the cost function that is the difference between the desired
output and the output by using the direct method that is stochastic
gradient descent or using the PSO iterative method in combination with
Simulated annealingiterative method to search efficiently and to be able
to converge to the global minimum.
 
This is the basic of Deep Learning, and this is what i have learned
as a software programmer to know more about artificial intelligence.
 
You can learn more about Deep learning here:
 
http://alexminnaar.com/deep-learning-basics-neural-networks-backpropagation-and-stochastic-gradient-descent.html
 
 
 
Thank you,
Amine Moulay Ramdane.
bleachbot <bleachbot@httrack.com>: Aug 25 10:44PM +0200

bleachbot <bleachbot@httrack.com>: Aug 25 11:02PM +0200

bleachbot <bleachbot@httrack.com>: Aug 26 01:15AM +0200

Mr Flibble <flibbleREMOVETHISBIT@i42.co.uk>: Aug 25 08:00PM +0100

On 25/08/2016 18:20, Gareth Owen wrote:
> improve cache behaviour. And since the compiler knows what the "normal"
> path through the code is, that can improve performance of branch
> prediction and speculative execution, as well as cache locality.
 
Indeed, try telling Jerry Stuckle that.
 
/Flibble
Ian Collins <ian-news@hotmail.com>: Aug 26 07:26AM +1200

On 08/26/16 01:30 AM, Jerry Stuckle wrote:
>> error returns for every function.
 
> Unwinding the stack via exception handling is much more CPU intensive
> than returning from functions.
 
Evidence?
 
--
Ian
legalize+jeeves@mail.xmission.com (Richard): Aug 25 08:01PM

[Please do not mail me a copy of your followup]
 
Ian Collins <ian-news@hotmail.com> spake the secret code
>> Unwinding the stack via exception handling is much more CPU intensive
>> than returning from functions.
 
>Evidence?
 
Every measured comparison I've seen says they come out about the
same.
 
Of course, if you ignore all the error return codes and don't
properly handle them, that is obviously fewer cycles since you're
comparing not doing error handling to doing error handling with
exceptions.
--
"The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline>
The Computer Graphics Museum <http://computergraphicsmuseum.org>
The Terminals Wiki <http://terminals.classiccmp.org>
Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>
David Brown <david.brown@hesbynett.no>: Aug 25 10:09PM +0200

On 25/08/16 19:13, Gareth Owen wrote:
> whenever you try and do something complicated.
 
> Consider a Matrix algebra package that can go wrong in multiple ways
> (singular matrices, numeric instability, incompatible dimensions etc).
 
In most cases, incompatible dimensions should be caught at compile time.
But the other failures could well happen.
 
 
> fail:
> return -EMATRIX_ERROR;
 
> No thanks.
 
I also say "no thanks" to a goto "solution". But there are many other
ways to handle this. For example, the Matrix class could hold a
information about the validity of its contents and possible errors -
just like NaN's in floating point. You can write your code that uses
Matrix's just the way you want, and at the end (or at any other
convenient points) check for safe results.
 
And yes, exceptions are a perfectly good way to handle this. I am not
saying exceptions don't have their use or their place - I am just saying
that they are not without their own costs, and are not always the best
way of handling things. In a situation like this, they could give a
very neat and clear way of handling local problems - precisely because
they can be treated locally rather than wandering off through innocent code.
Ian Collins <ian-news@hotmail.com>: Aug 26 08:11AM +1200

On 08/26/16 08:01 AM, Richard wrote:
 
>> Evidence?
 
> Every measured comparison I've seen says they come out about the
> same.
 
Same here along with the measurements I have made.
 
> properly handle them, that is obviously fewer cycles since you're
> comparing not doing error handling to doing error handling with
> exceptions.
 
That's right. Often in the case where exceptions are used but not
thrown (that is most of the time), using exceptions is slightly faster
than checking error codes.
 
--
Ian
Jerry Stuckle <jstucklex@attglobal.net>: Aug 25 04:47PM -0400

On 8/25/2016 3:26 PM, Ian Collins wrote:
 
>> Unwinding the stack via exception handling is much more CPU intensive
>> than returning from functions.
 
> Evidence?
 
Try it yourself and find out. But I know that is beyond the limited
capabilities of a troll.
 
 
--
==================
Remove the "x" from my email address
Jerry Stuckle
jstucklex@attglobal.net
==================
Jerry Stuckle <jstucklex@attglobal.net>: Aug 25 04:55PM -0400

On 8/25/2016 3:00 PM, Mr Flibble wrote:
>> prediction and speculative execution, as well as cache locality.
 
> Indeed, try telling Jerry Stuckle that.
 
> /Flibble
 
If you don't believe me, try believing Google. They have a good writeup
on exceptions in their style guide:
https://google.github.io/styleguide/cppguide.html#Exceptions
 
Although I wouldn't go so far as to say to "never use exceptions", I
would point out one statement in their discussion: "he availability of
exceptions may encourage developers to throw them when they are not
appropriate or recover from them when it's not safe to do so. For
example, invalid user input should not cause exceptions to be thrown."
 
Exactly the opposite of what you claimed.
 
 
--
==================
Remove the "x" from my email address
Jerry Stuckle
jstucklex@attglobal.net
==================
Mr Flibble <flibbleREMOVETHISBIT@i42.co.uk>: Aug 25 10:07PM +0100

On 25/08/2016 21:55, Jerry Stuckle wrote:
> appropriate or recover from them when it's not safe to do so. For
> example, invalid user input should not cause exceptions to be thrown."
 
> Exactly the opposite of what you claimed.
 
If you read the entire section of the Google style guide on exceptions
you would know that they require this prohibition for practical reasons
(to be compatible with their existing code-base). At the end of the
section Google say:
 
"Things would probably be different if we had to do it all over again
from scratch."
 
So assuming you are not writing code for a Google open source project
(most likely the case) then, no, the Google style guide is not a good
document to instruct one how to write modern C++ code and whether or not
to use exceptions.
 
So it seems your fractal wrongness happily continues unabated.
 
/Flibble
Ramine <ramine@1.1>: Aug 25 05:02PM -0400

Hello...........
 
Here is the remaining of my invention on artificial intelligence..
 
If you have read my previous post:
 
I have explained to you that we human feel the time, this is why
we can compose logic.
 
Now how can we feel the time ? we can feel the time because
we feel the space, so when we say for exemple an IF-Then statement
in logic, we are mapping in our mind the empirical experience
of our feelings of the space, that for exemple this object is before
this object, or this object is after this object, this is why we can
feel the IF-Then statement, and since we feel it this way, this is what
permit us to undertand what is logic and to understand the essence of
logic and this permit us to compose logic , my new invention in
artificial intelligence says so.
 
So i don't think that machines will be able to feel the time
and space like human are feeling time and space with there body
and minds, so i don't think that machines will be able to compose
logic such us human are composing it.
 
 
 
Thank you,
Amine Moulay Ramdane.
Ramine <ramine@1.1>: Aug 25 04:45PM -0400

Hello,
 
Here is my new invention on artificiel intelligence.
 
I have just read about Deep Learning and i have understood it,
please read here to understand more about it:
 
http://alexminnaar.com/deep-learning-basics-neural-networks-backpropagation-and-stochastic-gradient-descent.html
 
As you have noticed the philosophy of Deep Leaning is how to
control the weights and biases to minimize the error between
the desired output and the output by using gradient descent
with partial derivates or by using iterative methods such
us PSO.
 
But here is my invention:
 
The problem with artificiel intelligence is that scientists are thinking
that we can recognize high level objects and we can compose reasonning
by just programming it with machines using Deep Learning and softwares,
but this is false, i will give you an exemple:
 
When you say the water is fluid, you are not just saying it but
you are also feeling the fluidity of the water using the biology
of your body and mind.
 
Now i will make you understand more my invention in artificiel intelligence:
 
If you say a logical sentence such us:
 
If i enter the house, then i will stay in it for 6 hours.
 
The If-Then statement is how i am composing my logical reasonning,
but logical reasonning is not just saying this statement, logical
reasonning is feeling the variables that are space and time with your
body and mind, because when you say If-Then statement, you are feeling
with your body and mind that what is coming after the "Then" comes
in time after what comes after the "IF", so you are feeling the "after"
and "before" with you mind, so you are feeling the "time" with your
mind, and since we are feeling it like this with our mind we are
understanding the If-Then statement correctly, and this is what
permits the childs to understand correctly and compose logic,
such us logical inference.
 
So i don't think that machines will be able to feel the time
and space like human are feeling time and space with there body
and minds, so i don't think that machines will be able to compose
logic such us human are composing it.
 
 
 
Thank you,
Amine Moulay Ramdane.
Cholo Lennon <chololennon@hotmail.com>: Aug 25 05:36PM -0300

On 08/25/2016 08:10 AM, Bo Persson wrote:
 
> The language standard does not limit itself to file systems with a
> directory structure. z/OS used on IBM mainframes is an important example.
 
> So it might depend on exactly HOW portable you want to be.
 
Well, in my particular case I just need the same behavior on Windows,
Linux and Solaris, so I suppose my code is "portable" across those
platforms.
 
Regards
 
--
Cholo Lennon
Bs.As.
ARG
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to comp.lang.c+++unsubscribe@googlegroups.com.
Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)