Skip to content

Update sklearnex gpu to use array_api_dispatch and dpnp #218

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
ethanglaser wants to merge 1 commit into
base: main
Choose a base branch
from
Draft

Update sklearnex gpu to use array_api_dispatch and dpnp #218

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,6 +40,12 @@ conda env create -n sklearn -f envs/conda-env-sklearn.yml
conda env create -n rapids --solver=libmamba -f envs/conda-env-rapids.yml
```

GPU benchmarks using `sklearnex`'s `device: gpu` rely on scikit-learn's Array API support (`array_api_dispatch`), which requires `SCIPY_ARRAY_API=1` to be set in the environment before running:

```bash
export SCIPY_ARRAY_API=1
```

## 🚀 How To Use Scikit-learn_bench

### Benchmarks Runner
Expand Down
2 changes: 1 addition & 1 deletion configs/BENCH-CONFIG-SPEC.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -98,7 +98,7 @@ Configs have the three highest parameter keys:
| `data`:`distributed_split` | None | None, `rank_based` | Split type used to distribute data between machines in distributed algorithm. `None` type means usage of all data without split on all machines. `rank_based` type splits the data equally between machines with split sequence based on rank id from MPI. |
|<h3>Algorithm parameters</h3>||||
| `algorithm`:`library` | None | | Python module containing measured entity (class or function). |
| `algorithm`:`device` | `default` | `default`, `cpu`, `gpu` | Device selected for computation. |
| `algorithm`:`device` | `default` | `default`, `cpu`, `gpu` | Device selected for computation. `sklearnex`+`gpu` cases enable sklearn's `array_api_dispatch` and use `dpnp` data by default (see `sklearn_context` below); requires `SCIPY_ARRAY_API=1` to be set in the environment. |

## Benchmark-Specific Parameters

Expand Down
18 changes: 12 additions & 6 deletions configs/common/sklearn.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,12 +6,18 @@
{ "library": "sklearnex", "device": "cpu" }
]
},
"sklearn-ex[cpu,gpu] implementations": {
"algorithm": [
{ "library": "sklearn", "device": "cpu" },
{ "library": "sklearnex", "device": ["cpu", "gpu"] }
]
},
"sklearn-ex[cpu,gpu] implementations": [
{ "algorithm": { "library": "sklearn", "device": "cpu" } },
{ "algorithm": { "library": "sklearnex", "device": "cpu" } },
{
"algorithm": {
"library": "sklearnex",
"device": "gpu",
"sklearn_context": { "array_api_dispatch": true }
},
"data": { "format": "dpnp" }
}
],
"sklearnex spmd implementation": {
"algorithm": {
"library": "sklearnex.spmd",
Expand Down
Loading