cmedina/agenx-lora-training
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix: use torchrun for distributed training, load model on CPU

Browse Source
This commit is contained in:
Christian Medina
2026-06-30 18:45:14 -04:00
parent 5ad5518ca6
commit 75aeeeb20a
2 changed files with 5 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
train-on-this-server.sh
View File

@@ -68,7 +68,8 @@ echo "Training will take 6-24 hours depending on GPU."
echo "Press Ctrl+C to stop (model will be saved at checkpoint)."
echo ""
python3 training/scripts/train.py --config training/configs/llama2-7b-lora.yaml
# Use torchrun for distributed training (2 GPUs)
torchrun --nproc_per_node=2 training/scripts/train.py --config training/configs/llama2-7b-lora.yaml
echo ""
echo "=== Training Complete ==="