5,323
edits
PyTorch: Difference between revisions
→Debugging
| Line 121: | Line 121: | ||
<pre> | <pre> | ||
CUDA_LAUNCH_BLOCKING=1 python app.py | CUDA_LAUNCH_BLOCKING=1 python app.py | ||
</pre> | |||
For the following error: | |||
<pre> | |||
CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasGemmEx(...)` | |||
</pre> | |||
First check all your tensor types and shapes.<br> | |||
If you've checked all your tensor shapes and types and you can try running with the environment variable: | |||
<pre> | |||
CUBLAS_WORKSPACE_CONFIG=:0:0 | |||
</pre> | </pre> | ||