1 files changed, 59 insertions, 0 deletions
diff --git a/collaborativeagents/scripts/exp_vanilla.sbatch b/collaborativeagents/scripts/exp_vanilla.sbatch
new file mode 100644
index 0000000..445f771
--- /dev/null
+++ b/collaborativeagents/scripts/exp_vanilla.sbatch
@@ -0,0 +1,59 @@
+#!/bin/bash
+#SBATCH --job-name=exp_vanilla
+#SBATCH --account=bfqt-delta-gpu
+#SBATCH --partition=gpuH200x8
+#SBATCH --gres=gpu:4
+#SBATCH --nodes=1
+#SBATCH --ntasks=1
+#SBATCH --cpus-per-task=32
+#SBATCH --mem=256G
+#SBATCH --time=12:00:00
+#SBATCH --output=/projects/bfqt/users/yurenh2/ml-projects/personalization-user-model/exp_vanilla-%j.out
+#SBATCH --error=/projects/bfqt/users/yurenh2/ml-projects/personalization-user-model/exp_vanilla-%j.err
+
+set -e
+cd /projects/bfqt/users/yurenh2/ml-projects/personalization-user-model/collaborativeagents
+source /u/yurenh2/miniforge3/etc/profile.d/conda.sh
+conda activate eval
+
+export HF_HOME=/projects/bfqt/users/yurenh2/hf_cache/huggingface
+export PYTHONPATH="${PWD}:${PWD}/scripts:${PWD}/../src:${PYTHONPATH}"
+
+MODEL_8B="/projects/bfqt/users/yurenh2/ml-projects/personalization-user-model/models/llama-3.1-8b-instruct"
+PROFILE_PATH="/projects/bfqt/users/yurenh2/ml-projects/personalization-user-model/collaborativeagents/data/complex_profiles_v2/profiles_200.jsonl"
+
+echo "=== VANILLA (BATCH) ==="
+date
+nvidia-smi --query-gpu=index,name,memory.total --format=csv
+
+pkill -f "vllm.entrypoints" 2>/dev/null || true
+sleep 2
+
+CUDA_VISIBLE_DEVICES=0,1 python -m vllm.entrypoints.openai.api_server \
+    --model $MODEL_8B --port 8004 --tensor-parallel-size 2 \
+    --gpu-memory-utilization 0.90 --max-model-len 8192 \
+    --disable-log-requests --dtype bfloat16 &
+
+CUDA_VISIBLE_DEVICES=2,3 python -m vllm.entrypoints.openai.api_server \
+    --model $MODEL_8B --port 8003 --tensor-parallel-size 2 \
+    --gpu-memory-utilization 0.90 --max-model-len 8192 \
+    --disable-log-requests --dtype bfloat16 &
+
+echo "Waiting for servers..."
+for i in $(seq 1 200); do
+    if curl -s http://localhost:8004/health > /dev/null 2>&1 && curl -s http://localhost:8003/health > /dev/null 2>&1; then
+        echo "Servers ready after $((i*3))s"; break
+    fi
+    sleep 3
+done
+
+python scripts/run_experiments.py \
+    --methods vanilla \
+    --datasets math-hard,math-500,bigcodebench \
+    --n-profiles 200 --n-sessions 30 --max-turns 15 \
+    --use-vllm --batch-size 50 --parallel-profiles 50 \
+    --output-dir ../results/full_h200 \
+    --profile-path "$PROFILE_PATH"
+
+pkill -f "vllm.entrypoints" 2>/dev/null || true
+echo "Done: $(date)"