A scientific computing and data-handling library for C, providing functions on the same level as those of the typical stats package (OLS, probit, singular value decomposition, &c.). Written to facilitate writing fast and easy C code.
The project has moved to https://github.com/b-k/Apophenia .
- Easy handling of real-world data sets
- Description and manipulation of statistical (and all sorts of other) models
Easy to understand code. This project is well documented with an author of good writing style, but its state of development is a bit untidy at the moment (14_Jun_10 edition) and needs neatening up (IMHO). If you have the expertise in C and can understand the statistics this is a great resource.
The library seems to have devolved into the developer's own personal toolkit. The book/manual was written for some unknown version and the examples are completely broken on all of he latest releases. The library itself cannot even complete its own testing suite without segfaulting. The source code seems like yet another attempt to recreate C++ in C. The levels of complexity have probably contributed to the bugs and memory faults. A KISS (keep it simple...) approach might be a good idea here. I was looking for a solid C statistics library. Unfortunately I'll have to go with calling R from my C code. Apophenia would have been perfect. :-(
I use this daily at work. [I suppose as the lead developer, this is not surprising.]