Intel GPU Max Support + Gaudi (#214)

* Fix Whisper test by adding --num-workers to hf benchmark suite (#137) (#139) * Add brax benchmark * Update Metric measure * Add benchmate * Sync with stable --------- Co-authored-by: Pierre Delaunay <[email protected]> Co-authored-by: Olivier Breuleux <[email protected]>
mila-iqia · Jun 12, 2024 · 717de64 · 717de64
1 parent 702c1c6
commit 717de64
Show file tree

Hide file tree

Showing 150 changed files with 15,420 additions and 2,096 deletions.
diff --git a/.gitattributes b/.gitattributes
@@ -0,0 +1,2 @@
+requirements.*.txt linguist-generated
+.pin/*.txt linguist-generated
diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
@@ -88,6 +88,7 @@ jobs:
         run: |
           MILABENCH_GPU_ARCH=cuda  poetry run milabench pin -c constraints/cuda.txt --config config/standard.yaml 
           MILABENCH_GPU_ARCH=rocm  poetry run milabench pin -c constraints/rocm.txt --config config/standard.yaml 
+          MILABENCH_GPU_ARCH=xpu  poetry run milabench pin -c constraints/xpu.txt --config config/standard.yaml 
           git diff --stat
 
       - name: tests

diff --git a/.gitignore b/.gitignore
@@ -26,3 +26,20 @@ workspace/
 slurm-*
 scripts/inventory.yaml
 out.txt
+
+scripts/article/hpu/
+scripts/article/cuda/
+scripts/article/xpu/
+
+dependencies/
+benchmarks/gflownet/gflownet
+
+scripts/inventory.yaml
+output/
+sqlite.db
+.ruff_cache/
+
+test.out
+output/
+workspace/
+.pin/tmp-*
diff --git a/scripts/config.yaml → .no_report b/scripts/config.yaml → .no_report
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		requirements.*.txt linguist-generated
		.pin/*.txt linguist-generated