Skip to content

Commit

Permalink
Allow multiple OpenMP runtimes to be linked
Browse files Browse the repository at this point in the history
  • Loading branch information
tnunamak committed Mar 18, 2024
1 parent e252144 commit b70b758
Show file tree
Hide file tree
Showing 5 changed files with 241 additions and 1,386 deletions.
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -117,7 +117,7 @@ For most users, the easiest way to install Selfie is to follow the [Quick Start]
6. Run `poetry install` to install required Python dependencies.
7. Optional: Run `./scripts/llama-cpp-python-cublas.sh` to enable hardware acceleration (for details, see [Scripts](#llama-cpp-python-cublassh)).
8. Run `poetry run python -m selfie`, or `poetry run python -m selfie --gpu` if your device is GPU-enabled. The first time you run this, it will download ~4GB of model weights.
- On macOS, you may need to run `OMP_NUM_THREADS=1 poetry run python -m selfie` to avoid segmentation faults (with or without `--gpu`). [Read more here](https://github.com/vana-com/selfie/issues/33#issuecomment-2004637058).
- On macOS, you may need to run `OMP_NUM_THREADS=1 KMP_DUPLICATE_LIB_OK=TRUE poetry run python -m selfie` to avoid OpenMP errors (with or without `--gpu`). [Read more about OMP_NUM_THREADS here](https://github.com/vana-com/selfie/issues/33#issuecomment-2004637058).

[//]: # (1. `git clone
[//]: # (Disable this note about installing with GPU support until supported via transformers, etc.)
Expand Down
Loading

0 comments on commit b70b758

Please sign in to comment.