Hugging Face's logo

zhouyik
/
DenseLabelDev

TensorBoard
ONNX
Safetensors
Model card Files
xet
Metrics Community
DenseLabelDev / docs /MULTI_NODES.md
zhouyik's picture
zhouyik
Upload folder using huggingface_hub
032e687 verified
preview code
|
raw
history blame contribute delete
606 Bytes

IP 

hostname -I

Command 

# for example, 2 nodes

# on node 0
NNODES=2 NODE_RANK=0 PORT=3333 MASTER_ADDR=10.136.98.202 bash tools/dist.sh train projects/llava_sam2/configs/llava_sam2_internvl_v2_phi3_4b_llava_4ires_4vres_vchat_unfrozenSam2Dec.py 8 --deepspeed ./projects/llava_sam2/deepspeed_zero2_sam2.json

# on node 1
NNODES=2 NODE_RANK=1 PORT=3333 MASTER_ADDR=10.136.98.202 bash tools/dist.sh train projects/llava_sam2/configs/llava_sam2_internvl_v2_phi3_4b_llava_4ires_4vres_vchat_unfrozenSam2Dec.py 8 --deepspeed ./projects/llava_sam2/deepspeed_zero2_sam2.json