make guassian get rid of LAPACK #3819
Conversation
scripts/check_format.sh
Outdated
| @@ -53,4 +53,4 @@ if [[ $# -ne 2 ]]; then | |||
| fi | |||
|
|
|||
| # Run the check | |||
| check_shogun_style ${1:-} ${2:-} | |||
| check_shogun_style ${1:-} ${2:-} | |||
| CblasRowMajor, num_dim, num_dim, alpha_k[j], v.vector, 1, | ||
| v.vector, 1, (double*)cov_sum.matrix, num_dim); | ||
| #else | ||
| linalg::cblas_dger<float64_t>(alpha_k[j], v, v, cov_sum); |
There was a problem hiding this comment.
please use a different name, dont use cblas_dger
| cblas_dgemv(CblasRowMajor, CblasNoTrans, m_d.vlen, m_d.vlen, | ||
| 1, m_u.matrix, m_d.vlen, difference, 1, 0, temp_holder, 1); | ||
| #else | ||
| linalg::cblas_dgemv<float64_t>( |
There was a problem hiding this comment.
same here.... use a better name for the function cblas_dgemv
MikeLing
force-pushed
the
get_rid_of_LAPACK
branch
5 times, most recently
from
833e82a
to
4a894aa
Compare
| * @param vector y | ||
| */ | ||
| template <typename T> | ||
| void cblas_dgemv( |
There was a problem hiding this comment.
@vigsterkr how about vvpm which means vector produce vector then plus matrix . I know it still looks weird, but I can't find a better name to represent this function does like vector produce vector then plus matrix
There was a problem hiding this comment.
ok so let's just have it without the cblas_ prefix as @lisitsyn suggested... so dgemv
MikeLing
force-pushed
the
get_rid_of_LAPACK
branch
2 times, most recently
from
0b2fec2
to
c4007ea
Compare
|
Ask review for this pr. I guess it's ready to reviewed except the function names :) @vigsterkr Do you have time to take a look at this? Thank you! |
| @@ -350,6 +350,26 @@ Container<T> add(Container<T>& a, Container<T>& b, T alpha=1, T beta=1) | |||
| } | |||
|
|
|||
| /** | |||
| * Performs the operation A = alpha * x * y' + A | |||
| * | |||
| * @param scaling factor for vector x | |||
There was a problem hiding this comment.
you mean:
@param alpha scaling factor for vector x
| * and multiplies the resulting matrix by alpha. It then multiplies vector y by | ||
| * beta. It stores the sum of these two products in vector y | ||
| * | ||
| * @param scaling factor for vector ax |
There was a problem hiding this comment.
first word after the @param should be always the name of the variable in the function description
| * beta. It stores the sum of these two products in vector y | ||
| * | ||
| * @param scaling factor for vector ax | ||
| * @param vector a |
There was a problem hiding this comment.
isn't this a matrix? (2nd parameter of the function)
| * @param scaling factor for vector ax | ||
| * @param vector a | ||
| * @param Whether to transpose matrix a | ||
| * @param vector x |
There was a problem hiding this comment.
this should be
@param x vector
| * It then multiplies matrix c by beta. It stores the sum of these two products | ||
| * in matrix c. | ||
| * | ||
| * @param scaling factor for matrix a*b |
There was a problem hiding this comment.
same here as above... the doxygen needs to be fixed
MikeLing
force-pushed
the
get_rid_of_LAPACK
branch
from
c4007ea
to
85cfcca
Compare
|
@MikeLing cool let's see what the CI does. |
Those function names like 'cblas_dger' doesn't make sense to me.