feat:修改推荐补数据脚本

recommend/00_train_data_make.sh

@@ -1,7 +1,7 @@
 #!/bin/sh
 set -x
 
-dts=('20250222' '20250223' '20250224')
+dts=('20250225' '20250226' '20250227' '20250228' '20250301' '20250302' '20230303')
 for dt in "${dts[@]}"; do
     echo "开始处理: ${dt}"
 
@@ -36,13 +36,13 @@ for dt in "${dts[@]}"; do
     echo "${dt} 原始特征生产完成"
 
     /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_recsys.v20250218.makedata_recsys_41_data_fu_sample_20250218 \
-    --master yarn --driver-memory 6G --executor-memory 8G --executor-cores 1 --num-executors 30 \
+    --class com.aliyun.odps.spark.examples.makedata_recsys.v20250218.makedata_recsys_41_ros_train_filter_20250304 \
+    --master yarn --driver-memory 2G --executor-memory 5G --executor-cores 1 --num-executors 12 \
     --conf spark.driver.maxResultSize=2g \
     /root/zhaohp/recommend-emr-dataprocess/target/spark-examples-1.0.0-SNAPSHOT-shaded.jar \
     readPath:/dw/recommend/model/41_recsys_origin_date/${dt}*/* \
-    savePath:/dw/recommend/model/41_recsys_sample_data/${dt} \
-    fuSampleRate:1 whatLabel:is_share repartition:64
+    savePath:/dw/recommend/model/41_recsys_ros_train_data/${dt} \
+    whatLabel:is_share repartition:64
 
     echo "${dt} 负样本采样完成"
 done