Tadej Ciglarič
Tadej Ciglarič
Does single thread code using TBB exhibit the same performance as sequential code without TBB?
> What I'm thinking is something like That part looks more or less ok. I am just not sure if we want to do type conversions like that - Eigen...
> This gets weird because idk how to define Ret when a function can take in a matrix, vector, or row_vector. You would have to make separate overloads for these...
If the function is already exposed in stan language, the testing framework can get the list of signatures from the compiler. In that case you can run `runTests.py ./test/expressions --only-functions...
> @t4c1 should we have a specialized version of normal_id_glm() that works for vector ~ normal_id_glm(vector, real, real , real)? I doubt this would bring significant speedup over the code...
> To look at the indices when calling rvalue(...) and deduce if the user is asking for linear access like the above we could just call A.block(1, 1, 2, 2)...
I don't think so. Odd indeed.
I think none of these failures are actually related to the changes in this PR.
/verify with https://github.com/intel/llvm-test-suite/pull/1102
/verify with https://github.com/intel/llvm-test-suite/pull/1102