Genann VS ck

> I am very much a beginner in the space of machine learning

While the (precious and useful) advice around seem to cover mostly the bigger infrastructures, please note that

you can effectively do an important slice of machine learning work (study, personal research) with just a battery-efficiency-level CPU (not GPU), in the order of minutes, on a battery. That comes before going to "Big Data".

And there are lightweight tools: I am current enamoured with Genann («minimal, well-tested open-source library implementing feedfordward artificial neural networks (ANN) in C»), a single C file of 400 lines compiling to a 40kb object, yet well sufficient to solve a number of the problems you may meet.

https://codeplea.com/genann // https://github.com/codeplea/genann

After all, is it a good idea to have tools that automate process optimization while you are learning the deal? Only partially. You should build - in general and even metaphorically - the legitimacy of your Python ops on a good C ground.

And: note that you can also build ANNs in R (and other math or stats environments). If needed or comfortable...

Also note - reminder - that the MIT lessons of Prof. Patrick Winston for the Artificial Intelligence course (classical AI with a few lessons on ANNs) are freely available. That covers the grounds relative to climb into the newer techniques.

Genann - Simple ANN in C89, without additional dependencies. Zlib

Maybe I'm missing something, but x is not volatile and the compiler is free to assume that it is not modified concurrently outside the bounds of C's memory model. Compilers can and do hoist out loop invariants, and https://github.com/concurrencykit/ck/commit/b54ae5c4ace9b94442bbb46858449069f566d269 seems like an example of compilers doing what you say they don't. What am I missing?

Recommend checking out http://concurrencykit.org instead.

If you're interested in how other approaches work, or how one achieves concurrency on shared mutable state without mutual exclusion, would recommend checking out concurrency kit.

There are plenty of practical solutions to the safe memory reclamation problem in C. The language just doesn't force one on you.

From epoch-based reclamation (https://github.com/concurrencykit/ck/blob/master/include/ck_..., especially with the multiplexing extension to Fraser's classic scheme), to quiescence schemes (https://liburcu.org/), or hazard pointers (https://github.com/facebook/folly/blob/master/folly/synchron..., or https://pvk.ca/Blog/2020/07/07/flatter-wait-free-hazard-poin...)... or even simple using a type-stable (https://www.usenix.org/legacy/publications/library/proceedin...) memory allocator.

In my experience, it's easier to write code that is resilient to hiccups in C than in Java. Solving SMR with GC only offers something close to lock-freedom when you can guarantee global GC pauses are short enough... and common techniques to bound pauses, like explicitly managed freelists land you back in the same problem space as C.

ck - Concurrency primitives, safe memory reclamation mechanisms and non-blocking data structures. BSD-2-Clause

Indeed they do, https://github.com/concurrencykit/ck