startup.sh

#!/bin/bash

export HF_HOME="/data/.huggingface"

echo "PWD: $(pwd)"
echo $HF_TOKEN > .hf_token
echo "LS: $(ls -als)"

while true; do nvidia-smi; sleep 600; done &

python -c "import torch; \
    print(f\"is availeble = {torch.cuda.is_available()}\"); \
    print(f\"device count = {torch.cuda.device_count()}\"); \
    print(f\"current device = {torch.cuda.current_device()}\")"

python -m serve.controller --host 0.0.0.0 --port 10000 &
P1=$!

sleep 30

python -m serve.gradio_web_server --controller http://127.0.0.1:10000 --model-list-mode reload --share &
P2=$!

python -m interactive_demo --port 40000 --model_id prism-dinosiglip+7b &
P3=$!
# python -m interactive_demo --port 40001 --model_family llava-v15 --model_id llava-v1.5-7b --model_dir liuhaotian/llava-v1.5-7b &
# P4=$!

ls -als $HF_HOME
tree --du -h $HF_HOME


wait $P1 $P2 $P3 
# $P4