DouDou commited on
Upload data3/monitor_batch.sh with huggingface_hub
Browse files- data3/monitor_batch.sh +55 -0
data3/monitor_batch.sh
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
#!/bin/bash
|
| 2 |
+
# 监控批处理任务状态
|
| 3 |
+
|
| 4 |
+
set -e
|
| 5 |
+
|
| 6 |
+
BATCH_ID_FILE="batch_id.txt"
|
| 7 |
+
|
| 8 |
+
# 获取 Batch ID
|
| 9 |
+
if [ -n "$1" ]; then
|
| 10 |
+
BATCH_ID=$1
|
| 11 |
+
elif [ -f "$BATCH_ID_FILE" ]; then
|
| 12 |
+
BATCH_ID=$(cat $BATCH_ID_FILE)
|
| 13 |
+
else
|
| 14 |
+
echo "❌ 错误: 请提供 Batch ID"
|
| 15 |
+
echo "用法: $0 <batch_id>"
|
| 16 |
+
echo "或确保 batch_id.txt 文件存在"
|
| 17 |
+
exit 1
|
| 18 |
+
fi
|
| 19 |
+
|
| 20 |
+
echo "🔍 监控 Batch 任务: $BATCH_ID"
|
| 21 |
+
echo "按 Ctrl+C 退出监控"
|
| 22 |
+
echo "========================================"
|
| 23 |
+
|
| 24 |
+
while true; do
|
| 25 |
+
TIMESTAMP=$(date '+%Y-%m-%d %H:%M:%S')
|
| 26 |
+
echo ""
|
| 27 |
+
echo "[$TIMESTAMP]"
|
| 28 |
+
|
| 29 |
+
STATUS_OUTPUT=$(python3 generate_problems_batch.py status $BATCH_ID)
|
| 30 |
+
echo "$STATUS_OUTPUT"
|
| 31 |
+
|
| 32 |
+
# 检查是否完成
|
| 33 |
+
if echo "$STATUS_OUTPUT" | grep -q "Status: completed"; then
|
| 34 |
+
echo ""
|
| 35 |
+
echo "✅ 批处理已完成!"
|
| 36 |
+
echo ""
|
| 37 |
+
echo "下载结果:"
|
| 38 |
+
echo " python3 generate_problems_batch.py download $BATCH_ID --output batch_results_raw.jsonl"
|
| 39 |
+
echo ""
|
| 40 |
+
echo "处理结果:"
|
| 41 |
+
echo " python3 generate_problems_batch.py process --input batch_results_raw.jsonl --output programming_problems_batch.jsonl"
|
| 42 |
+
break
|
| 43 |
+
elif echo "$STATUS_OUTPUT" | grep -q "Status: failed"; then
|
| 44 |
+
echo ""
|
| 45 |
+
echo "❌ 批处理失败!"
|
| 46 |
+
break
|
| 47 |
+
elif echo "$STATUS_OUTPUT" | grep -q "Status: expired"; then
|
| 48 |
+
echo ""
|
| 49 |
+
echo "❌ 批处理已过期!"
|
| 50 |
+
break
|
| 51 |
+
fi
|
| 52 |
+
|
| 53 |
+
echo "⏳ 等待 5 分钟后再次检查..."
|
| 54 |
+
sleep 300
|
| 55 |
+
done
|