Alexey Solovyev
Alexey Solovyev
The `coq_codewars` code for checking assumptions is taken directly from the `Print Assumptions` implementation (there is a comment about this). So I would say that correctly working `Print Assumptions` is...
Nothing is wrong in your example. But it cannot be analyzed by FPTaylor because the expression under the square root can be 0. FPTaylor needs to find derivatives to compute...
Another way to solve your compilation issue is to install the legacy `Num` library for OCaml 4.06: `opam install num`. I will update the Readme file with this info. I...
Error bounds are printed as hexadecimal numerals now and the minimal OCaml version is increased to 4.03.
FPTaylor always assumes that errors are symmetric, i.e., if `X` is real then `Round(X) in [X - e, X + e]`. This estimate can be improved because errors are monotone:...
@zvonimir Comments are always good. So please create a pull request and I will review and merge it. @ganeshutah There is [`ocamldoc`](http://caml.inria.fr/pub/docs/manual-ocaml/ocamldoc.html) for generating html and LaTeX files from OCaml...
I have implemented custom floating-point formats in the new branch [rounding](https://github.com/soarlab/FPTaylor/tree/rounding). Here is a simple example: ``` Variables // = 5 significand bits (including the implicit bit) and 2 exponent...
Please delete all binary files from this PR (*.cmo, *.cmi, etc.)
Everything works well when the cost function is `sqrt(x1)`
Ranges [-pi/2,pi/2] for arcsin and [0,pi] for arccos are most common so I agree with Gaol choices.