Can normal_id_glm use a GPU when the design matrix is not fixed?

Modeling
3

I hadn’t tried that. When X2 is built in the transformed data block, the utilization picks up (e.g., below). Though, at least in my application this won’t work since X needs to be built from model parameters.

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 418.130      Driver Version: 418.130      CUDA Version: 10.1     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GRID P100-8Q        On   | 00000000:02:04.0 Off |                  N/A |
| N/A   N/A    P0    N/A /  N/A |    819MiB /  8192MiB |     28%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0     24541      C   ./normal_glm                                 291MiB |
+-----------------------------------------------------------------------------+