recommend-emr-dataprocess/qiaojialiang/handle_rov.sh

#!/bin/sh
set -ex

source /root/anaconda3/bin/activate py37

#  nohup sh handle_rov.sh > "$(date +%Y%m%d_%H%M%S)_handle_rov.log" 2>&1 &

# 原始数据table name
table='alg_recsys_sample_all'
today="$(date +%Y%m%d)"
today_early_3="$(date -d '3 days ago' +%Y%m%d)"
#table='alg_recsys_sample_all_test'
# 处理分区配置 推荐数据间隔一天生产，所以5日0点使用3日0-23点数据生产new模型数据
begin_early_2_Str="$(date -d '2 days ago' +%Y%m%d)"
end_early_2_Str="$(date -d '2 days ago' +%Y%m%d)"
beginHhStr=00
endHhStr=23
max_hour=05
max_minute=00
# 各节点产出hdfs文件绝对路径
# 源数据文件
originDataPath=/dw/recommend/model/13_sample_data/
# 特征值
valueDataPath=/dw/recommend/model/14_feature_data/
# 特征分桶
bucketDataPath=/dw/recommend/model/16_train_data/
# 模型数据路径
MODEL_PATH=/root/joe/recommend-emr-dataprocess/model
# 预测路径
PREDICT_PATH=/root/joe/recommend-emr-dataprocess/predict
# 历史线上正在使用的模型数据路径
LAST_MODEL_HOME=/root/joe/model_online
# 模型数据文件前缀
model_name=aka8
# fm模型
FM_HOME=/root/sunmingze/alphaFM/bin
# hadoop
HADOOP=/opt/apps/HADOOP-COMMON/hadoop-common-current/bin/hadoop
OSS_PATH=oss://art-recommend.oss-cn-hangzhou.aliyuncs.com/zhangbo/



# 0 对比AUC 前置对比2日模型数据 与 线上模型数据效果对比，如果2日模型优于线上，更新线上模型
echo "$(date +%Y-%m-%d_%H-%M-%S)----------step0------------开始对比，新：${MODEL_PATH}/${model_name}_${today_early_3}.txt，与线上online模型数据auc效果"
$HADOOP fs -text ${bucketDataPath}/${begin_early_2_Str}/* | ${FM_HOME}/fm_predict -m ${LAST_MODEL_HOME}/model_online.txt -dim 8 -core 8 -out ${PREDICT_PATH}/${model_name}_${today}_online.txt
$HADOOP fs -text ${bucketDataPath}/${begin_early_2_Str}/* | ${FM_HOME}/fm_predict -m ${MODEL_PATH}/${model_name}_${today_early_3}.txt -dim 8 -core 8 -out ${PREDICT_PATH}/${model_name}_${today}_new.txt

online_auc=`cat ${PREDICT_PATH}/${model_name}_${today}_online.txt | /root/sunmingze/AUC/AUC`
if [ $? -ne 0 ]; then
   echo "推荐线上模型AUC计算失败"
#   /root/anaconda3/bin/python ad/ad_monitor_util.py "线上模型AUC计算失败"
   exit 1
fi

new_auc=`cat ${PREDICT_PATH}/${model_name}_${today}_new.txt | /root/sunmingze/AUC/AUC`
if [ $? -ne 0 ]; then
   echo "推荐新模型AUC计算失败"
#   /root/anaconda3/bin/python ad/ad_monitor_util.py "新模型AUC计算失败"
   exit 1
fi


# 1 对比auc数据判断是否更新线上模型
if [ "$online_auc" \< "$new_auc" ]; then
    echo "新模型优于线上模型: 线上模型AUC: ${online_auc}, 新模型AUC: ${new_auc}"
    # 1.1 模型格式转换
    cat ${MODEL_PATH}/${model_name}_${today_early_3}.txt |
    awk -F " " '{
        if (NR == 1) {
            print $1"\t"$2
        } else {
            split($0, fields, " ");
            OFS="\t";
            line="" 1; i <= 10 && i <= length(fields); i++) {
                line
            for (i = = (line ? line "\t" : "") fields[i];
            }
            print line
        }
    }' > ${MODEL_PATH}/${model_name}_${today_early_3}_change.txt
    if [ $? -ne 0 ]; then
       echo "新模型文件格式转换失败"
#       /root/anaconda3/bin/python ad/ad_monitor_util.py "新模型文件格式转换失败"
       exit 1
    fi
    # 1.2 模型文件上传OSS
    online_model_path=${OSS_PATH}/${model_name}.txt
    $HADOOP fs -test -e ${online_model_path}
    if [ $? -eq 0 ]; then
        echo "数据存在, 先删除。"
        $HADOOP fs -rm -r -skipTrash ${online_model_path}
    else
        echo "数据不存在"
    fi

    $HADOOP fs -put ${MODEL_PATH}/${model_name}_${today_early_3}_change.txt ${online_model_path}
    if [ $? -eq 0 ]; then
       echo "推荐模型文件至OSS成功"
    else
       echo "推荐模型文件至OSS失败"
       exit 1
    fi
    # 1.3 本地保存最新的线上使用的模型，用于下一次的AUC验证
    cp -f ${LAST_MODEL_HOME}/model_online.txt ${LAST_MODEL_HOME}/model_online_$(date +\%Y\%m\%d).txt
    cp -f ${MODEL_PATH}/${model_name}_${today_early_3}.txt ${LAST_MODEL_HOME}/model_online.txt
    if [ $? -ne 0 ]; then
       echo "模型备份失败"
#       /root/anaconda3/bin/python ad/ad_monitor_util.py "模型备份失败 - 最新模型地址: ${MODEL_PATH}/${model_name}_${today_early_1}.txt"
       exit 1
    fi
#    /root/anaconda3/bin/python ad/ad_monitor_util.py "新模型优于线上模型: 线上模型AUC: ${online_auc}, 新模型AUC: ${new_auc}"
else
    echo "新模型不如线上模型: 线上模型AUC: ${online_auc}, 新模型AUC: ${new_auc}"
#    /root/anaconda3/bin/python ad/ad_monitor_util.py "新模型不如线上模型: 线上模型AUC: ${online_auc}, 新模型AUC: ${new_auc}"
fi



# 2 判断上游表是否生产完成，最长等待到max_hour点
# shellcheck disable=SC2154
echo "$(date +%Y-%m-%d_%H-%M-%S)----------step1------------开始校验是否生产完数据，分区信息：beginStr:${begin_early_2_Str}${beginHhStr},endStr:${end_early_2_Str}${endHhStr}"
while true; do
  python_return_code=$(python /root/joe/recommend-emr-dataprocess/qiaojialiang/checkHiveDataUtil.py --table ${table} --beginStr ${begin_early_2_Str}${beginHhStr} --endStr ${end_early_2_Str}${endHhStr})
  echo "python 返回值：${python_return_code}"
  if [ $python_return_code -eq 0 ]; then
    echo "Python程序返回0，校验存在数据，退出循环。"
    break
  fi
  echo "Python程序返回非0值，不存在数据，等待五分钟后再次调用。"
  sleep 300
  current_hour=$(date +%H)
  current_minute=$(date +%M)
  # shellcheck disable=SC2039
  if (( current_hour > max_hour || (current_hour == max_hour && current_minute >= max_minute) )); then
    echo "最长等待时间已到，失败:${current_hour}-${current_minute}"
    exit 1
  fi
done

# 3 生产原始数据
echo "$(date +%Y-%m-%d_%H-%M-%S)----------step2------------开始根据${table}生产原始数据"
/opt/apps/SPARK2/spark-2.4.8-hadoop3.2-1.0.8/bin/spark-class2 org.apache.spark.deploy.SparkSubmit \
--class com.aliyun.odps.spark.examples.makedata_qiao.makedata_13_originData_20240705 \
--master yarn --driver-memory 1G --executor-memory 2G --executor-cores 1 --num-executors 16 \
../target/spark-examples-1.0.0-SNAPSHOT-shaded.jar \
tablePart:64 repartition:32 \
beginStr:${begin_early_2_Str}${beginHhStr} endStr:${end_early_2_Str}${endHhStr} \
savePath:${originDataPath} \
table:${table}
if [ $? -ne 0 ]; then
   echo "Spark原始样本生产任务执行失败"
   exit 1
else
   echo "spark原始样本生产执行成功"
fi


# 4 特征值拼接
echo "$(date +%Y-%m-%d_%H-%M-%S)----------step3------------开始特征值拼接"
/opt/apps/SPARK2/spark-2.4.8-hadoop3.2-1.0.8/bin/spark-class2 org.apache.spark.deploy.SparkSubmit \
--class com.aliyun.odps.spark.examples.makedata_qiao.makedata_14_valueData_20240705 \
--master yarn --driver-memory 1G --executor-memory 3G --executor-cores 1 --num-executors 32 \
../target/spark-examples-1.0.0-SNAPSHOT-shaded.jar \
readPath:${originDataPath} \
savePath:${valueDataPath} \
beginStr:${begin_early_2_Str} endStr:${end_early_2_Str} repartition:1000
if [ $? -ne 0 ]; then
   echo "Spark特征值拼接处理任务执行失败"
   exit 1
else
   echo "spark特征值拼接处理执行成功"
fi

# 5 特征分桶
echo "$(date +%Y-%m-%d_%H-%M-%S)----------step4------------根据特征分桶生产重打分特征数据"
/opt/apps/SPARK2/spark-2.4.8-hadoop3.2-1.0.8/bin/spark-class2 org.apache.spark.deploy.SparkSubmit \
--class com.aliyun.odps.spark.examples.makedata_qiao.makedata_16_bucketData_20240705 \
--master yarn --driver-memory 2G --executor-memory 4G --executor-cores 1 --num-executors 16 \
../target/spark-examples-1.0.0-SNAPSHOT-shaded.jar \
readPath:${valueDataPath} \
savePath:${bucketDataPath} \
beginStr:${begin_early_2_Str} endStr:${end_early_2_Str} repartition:1000
if [ $? -ne 0 ]; then
   echo "Spark特征分桶处理任务执行失败"
   exit 1
else
   echo "spark特征分桶处理执行成功"
fi

# 6 模型训练
echo "$(date +%Y-%m-%d_%H-%M-%S)----------step5------------开始模型训练"
$HADOOP fs -text ${bucketDataPath}/${begin_early_2_Str}/* | ${FM_HOME}/fm_train -m ${MODEL_PATH}/${model_name}_${begin_early_2_Str}.txt -dim 1,1,8 -im ${LAST_MODEL_HOME}/model_online.txt -core 8
if [ $? -ne 0 ]; then
   echo "模型训练失败"
#   /root/anaconda3/bin/python ad/ad_monitor_util.py "推荐模型训练失败"
   exit 1
fi

echo "$(date +%Y-%m-%d_%H-%M-%S)----------step6------------模型训练完成：${MODEL_PATH}/${model_name}_${begin_early_2_Str}.txt"