Convert from real12 to compilable real precision

nedtaylor / athena

A Fortran-based feed-forward neural network library. Whilst this library currently has a focus on 3D convolutional neural networks (CNNs), it can handle most standard hidden layer forms of neural networks, with the plan to integrate more.

MIT License

18 stars 2 forks source link

Convert from real12 to compilable real precision #36

Open nedtaylor opened 6 months ago

nedtaylor commented 6 months ago

This will remove the need for users to use real12 types, improving compatibility.

This will also allow users on compilation choose precision, and make it more inline with industry standards.

nedtaylor commented 3 months ago

Disagree with this. From discussions in various posts linked below, it appears that selected_XX_kind is the more portable and preferable option.

Simply change name of real12 to real32 for consistency.

https://github.com/fortran-lang/stdlib/issues/25 https://stevelionel.com/drfortran/2017/03/27/doctor-fortran-in-it-takes-all-kinds/