recommend-emr-dataprocess/ad_display/pipline/train_xgb_model.sh

#!/bin/sh

start_date=""
end_date=""
feature_file=""
if(($#==3))
then
    start_date=$1
    end_date=$2
    feature_file=$3
else
    exit -1
fi

# env
export HADOOP_CONF_DIR=/etc/taihao-apps/hadoop-conf
export JAVA_HOME=/usr/lib/jvm/java-1.8.0

# params
workers=16
minCnt=0
BASE_TRAIN_DATA_PATH=/dw/recommend/model/ad_display/sample
MODEL_SAVE_PATH=/dw/recommend/model/ad_display/model/model_xgb

# params
train_data_path=""
for((i=0; i<=21; i++))
do
  data_date=$(date -d "$start_date $i day" +"%Y%m%d")
  if [ "$data_date" -le "$end_date" ]
  then
    one_day_data_path="${BASE_TRAIN_DATA_PATH}/${data_date}"
    if [[ -z $train_data_path ]]
    then
        train_data_path=$one_day_data_path
    else
        train_data_path="$train_data_path,$one_day_data_path"
    fi
  fi
done

## ******* train *******
set -x
/opt/apps/SPARK3/spark-3.3.1-hadoop3.2-1.0.5/bin/spark-class org.apache.spark.deploy.SparkSubmit \
--class com.tzld.piaoquan.recommend.model.train_profile_gender_xgb_20251114 \
--master yarn --driver-memory 6G --executor-memory 10G --executor-cores 1 --num-executors ${workers} \
--conf spark.yarn.executor.memoryoverhead=2048 \
--conf spark.shuffle.service.enabled=true \
--conf spark.shuffle.service.port=7337 \
--conf spark.shuffle.consolidateFiles=true \
--conf spark.shuffle.manager=sort \
--conf spark.storage.memoryFraction=0.4 \
--conf spark.shuffle.memoryFraction=0.5 \
--conf spark.default.parallelism=200 \
--conf spark.sql.debug.maxToStringFields=100 \
--files ${feature_file} \
/mnt/disk1/jch/recommend-model/recommend-model-produce/target/recommend-model-produce-jar-with-dependencies.jar \
trainPath:${train_data_path} \
featureFile:${feature_file} \
minCnt:${minCnt} \
modelPath:${MODEL_SAVE_PATH} \
eta:0.06 gamma:0.0 max_depth:4 num_round:1000 num_worker:${workers}