Basic neural nets are like a 'bag of points' arranged in layers. each point has a 'weight' that takes the input, applies math using that weight to get an output. (It learns by adjusting the weight so that the 'inputs' will calculate to 'expected' outputs.
The next stage is called 'LSTM' (Long-Short Term Memory), this is like the 'bag of points' but has a 'Long memory' gate, a 'short memory' gate, and a 'forget' gate. So, when you type a phrase, it looks at the words before and tries to guess at what's coming next and will 'forget' options so that it doesn't go back to the previous guess.
Transformers take the whole document and create 'key-value' pairs that are also trained like the 'bag of points'. the benefit of this is that the 'transformer' can begin to learn the context based on the other features. This is a tricky one to explain, but is the first type of AI system that's capable of grasping context of what is 'scanned'.
To be fair, you can ignore all the silly stuff that has been piled on to C++ and just write C code. It will run on C++.
As much as I dislike Microsoft, I think C# is the easiest to use. No pointers, great framework with anything you could want, no include file nightmares. No macro nightmares. Garbage collection. Really good documentation. It's like playing a game with the difficulty set to the easiest setting.
C, C++, Java, C#, Pascal all use pretty similar syntax.
C# is closer to Java. Both C# and Java compile into CPU independent bytecode. Then at runtime the bytecode is interpreted, whereas in C, C++ you compile directly in the CPU's machine code.
I use C# every day. I think it is the easiest to use, but if you raw performance, then C or C++ because it is closest to the metal. If you really want max performance then assembler.
When I first learned c/c++ everything was all about OOP programming, it wasn't until later that I tried making a game using OOP, I had everything going great and then added an inherited additional behavior and suddenly I learned the hard way the flaw in OOP thinking.
c# is really the easiest, but that's because like Java, there's the code you write and then it adds piles of code to 'fill the gaps' to do the background work. The benefit of having the garbage collector do the heavy lifting, until you need a responsive app that suddenly calls the garbage collector when you don't want it and the whole program stops for a moment.
Even though there's more to worry about when handling memory, that greater level of control has value.
Yes I have run into garbage collector hiccups in Java. At the time I got around it by basically keeping a reference to the objects around when I was done with them, and then re-initialized them and re-used them as needed. So the garbage collector had nothing to do.
It was a little awkward but it worked great. I could get away with that because they were all fixed size.
I have to agree, the C# GC must be far more efficient, because with Java, especially on an android, when it hits you might lose a half-second (I know that sounds like a 'who cares', but it can cause grief)
I have no idea what he's talking about.
He's talking about Neural Nets / AI.
Basic neural nets are like a 'bag of points' arranged in layers. each point has a 'weight' that takes the input, applies math using that weight to get an output. (It learns by adjusting the weight so that the 'inputs' will calculate to 'expected' outputs.
The next stage is called 'LSTM' (Long-Short Term Memory), this is like the 'bag of points' but has a 'Long memory' gate, a 'short memory' gate, and a 'forget' gate. So, when you type a phrase, it looks at the words before and tries to guess at what's coming next and will 'forget' options so that it doesn't go back to the previous guess.
Transformers take the whole document and create 'key-value' pairs that are also trained like the 'bag of points'. the benefit of this is that the 'transformer' can begin to learn the context based on the other features. This is a tricky one to explain, but is the first type of AI system that's capable of grasping context of what is 'scanned'.
I too love C and always despised C++.
To be fair, you can ignore all the silly stuff that has been piled on to C++ and just write C code. It will run on C++.
As much as I dislike Microsoft, I think C# is the easiest to use. No pointers, great framework with anything you could want, no include file nightmares. No macro nightmares. Garbage collection. Really good documentation. It's like playing a game with the difficulty set to the easiest setting.
I never got beyond 6502 assembler. (Am I showing my age?)
OSI Challeger 1P
https://www.old-computers.com/museum/computer.asp?st=1&c=813
C, C++, Java, C#, Pascal all use pretty similar syntax.
C# is closer to Java. Both C# and Java compile into CPU independent bytecode. Then at runtime the bytecode is interpreted, whereas in C, C++ you compile directly in the CPU's machine code.
I use C# every day. I think it is the easiest to use, but if you raw performance, then C or C++ because it is closest to the metal. If you really want max performance then assembler.
When I first learned c/c++ everything was all about OOP programming, it wasn't until later that I tried making a game using OOP, I had everything going great and then added an inherited additional behavior and suddenly I learned the hard way the flaw in OOP thinking.
c# is really the easiest, but that's because like Java, there's the code you write and then it adds piles of code to 'fill the gaps' to do the background work. The benefit of having the garbage collector do the heavy lifting, until you need a responsive app that suddenly calls the garbage collector when you don't want it and the whole program stops for a moment.
Even though there's more to worry about when handling memory, that greater level of control has value.
Yes I have run into garbage collector hiccups in Java. At the time I got around it by basically keeping a reference to the objects around when I was done with them, and then re-initialized them and re-used them as needed. So the garbage collector had nothing to do.
It was a little awkward but it worked great. I could get away with that because they were all fixed size.
I've never yet had an issue with the GC with C#.
I have to agree, the C# GC must be far more efficient, because with Java, especially on an android, when it hits you might lose a half-second (I know that sounds like a 'who cares', but it can cause grief)
Only a select group of people knows what that even means. It’s not suppressed.
That’s just a bunch of Ching Chong bing bong to me, I wouldn’t like that either
Tbh that post went over the heads of many so I'm not surprised.
There are a lot of bots. All influencers have lots of them.