美购精排模型

spark/featureEng.py

@@ -662,11 +662,15 @@ if __name__ == '__main__':
     print("user feature to redis...")
     userDatas = collectFeaturesToDict(samplesWithUserFeatures, user_columns, "user")
     featureToRedis(FEATURE_USER_KEY, userDatas)
+    timestmp5 = int(round(time.time()))
+    print("user feature to redis 耗时s:{}".format(timestmp5 - timestmp4))
 
     # item特征数据存入redis
     print("item feature to redis...")
     itemDatas = collectFeaturesToDict(samplesWithUserFeatures, item_columns, "item")
     featureToRedis(FEATURE_ITEM_KEY, itemDatas)
+    timestmp6 = int(round(time.time()))
+    print("item feature to redis 耗时s:{}".format(timestmp6 - timestmp5))
 
     # model columns
     print("model columns to redis...")
@@ -679,5 +683,7 @@ if __name__ == '__main__':
     splitTimestamp = int(time.mktime(time.strptime(endDay, "%Y%m%d")))
     splitAndSaveTrainingTestSamplesByTimeStamp(samplesWithUserFeatures, splitTimestamp, TRAIN_FILE_PATH)
     print("write to hdfs success...")
+    timestmp7 = int(round(time.time()))
+    print("总耗时s:{}".format(timestmp7 - timestmp6))
 
     spark.stop()
\ No newline at end of file