Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
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
21 changes: 21 additions & 0 deletions multinode-example/task.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,21 @@
name: multinode-example
resources:
num_nodes: 2
cpus: 2
setup: |
python -V || true
echo "setup complete"
run: |
echo "=== Host/network context ==="
hostname
python - <<'PY'
import os, socket
keys = [
"MASTER_ADDR", "MASTER_PORT", "WORLD_SIZE",
"RANK", "LOCAL_RANK", "NODE_RANK"]
print("hostname:", socket.gethostname())
for k in keys:
print(f"{k}={os.environ.get(k)}")
PY

echo "done"
2 changes: 1 addition & 1 deletion video-diffusion/task.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,6 @@ resources:
accelerators: "A100:1"
envs:
PYTHONUNBUFFERED: "1"
HF_TOKEN: "ENTER_HF_TOKEN_HERE"
HF_TOKEN: "{{secret._HF_TOKEN}}"
setup: "uv pip install --upgrade pip && uv pip install git+https://github.com/huggingface/diffusers.git && uv pip install 'transformers>=4.41.0' 'peft>=0.17.0' accelerate ftfy huggingface-hub opencv-python transformerlab imageio-ffmpeg"
run: "python ~/video-diffusion/main.py"