This blog, and many others, are hosted using Amazon AWS. This one has a difference, it is almost completely free to host (and was completely free for the first year) because of Hugo, Amazon S3, Cloudfront, lambda@edge, Azure DevOps and (optionally) GitHub.
Microsoft recently announced .NET 5 (dotnet5) which is the unification of .NET Framework and dotnet core, leaving behind the divergance, and unifying the whole .NET platform.
As of C# 7.1, it is possible with console apps, to have async all the way to the entry point. The previous constraints of the entry point have been the same all the way up to this point, and similar to the entry point in C/C++ apps. The C# entry point method must be static, the name of the method must be Main, the return type must be either void or int and finally, the method can accept zero arguments, or exactly one argument of string[], which contains the command arguments.
In C# and in .NET in general, there is a generic stack class which accepts a single type parameter. The stack class is a FILO (first in last out) collection, and can be liked to stacking plates, you cannot remove the bottom plate, without first removing all the plates above it. The problem with the Stack<T> class is that it has an IEnumerable<T> constructor…
The C++ library (source on github) is designed to be a core neural network library, implementing basic neurons, layers and networks and comes with some basic learning methods too. The library is complete enough for those who want to learn and understand neural networks and how they are put together, but by no means is intended to be a complete AI library.
This is a short introduction into Dependency Injection for those who are new to dotnet core and C#, with an attempt to explain what it is, and how it works.
Applications feel more optimised when their binaries, or set of binaries are small. With dotnet core 3.0 there are some features built in that help facilitate this.