-
Hi, there!
First, I show you my example.
```
BLAS.set_num_threads(1)
block_mat1 = Matrix(1,1)
block_mat2 = Matrix(2,2)
block_mat3 = Matrix(3,3)
block_mat4 = Matrix(4,4)
block_mat1[1,1] = ones(…
-
See #355 for the background. I am now applying the same methodology to the matrix multiplication benchmark:
* https://juliasimd.github.io/LoopVectorization.jl/stable/examples/matrix_multiplication/…
-
Hi,
When I try to compile a matrix multiplication A = B*C where A, B and C all three are CSR, taco doesn't seem to produce the correct result.
```C
#include
#include "taco.h"
using namespa…
-
-
Following [a blog post on using Apple's BLAS library](https://blogs.mathworks.com/matlab/2023/12/13/life-in-the-fast-lane-making-matlab-even-faster-on-apple-silicon-with-apple-accelerate/) you can swi…
-
Hello!
Thank you for development of MOFA.
In my work, I wanted to use MOFA latent factors to predict a phenotype.
To prevent data leakage between my train and test set, I wanted to train the M…
-
```
Matrix multiplication that uses BLAS only supports the datatype: double.
It should be relatively easy to implement all BLAS supported datatypes:
float, double, complex, double complex.
```
…
-
### 🐛 Describe the bug
When running a .pte compile core ml model I am facing the following error:
`E 00:00:15.438547 executorch:method.cpp:69] Backend CoreMLBackend is not registered.`
Here is …
-
```
CSR * CSR on the device should not duplicate the column_indices and values
arrays to use the COO * COO implementation
```
Original issue reported on code.google.com by `wnbell` on 6 Jul 2011 at …
-
```
CSR * CSR on the device should not duplicate the column_indices and values
arrays to use the COO * COO implementation
```
Original issue reported on code.google.com by `wnbell` on 6 Jul 2011 at …