修改测试文件

tensnsorflow/multi.py

@@ -197,14 +197,16 @@ def con_sql(db,sql):
 
 
 def test():
+    sql = "select stat_date,cid_id,y,ccity_name from esmm_train_data limit 60"
+    rdd = spark.sql(sql).select("stat_date","cid_id","y","ccity_name").rdd
+    spark.createDataFrame(rdd).show(6)
+
     from hdfs import InsecureClient
     from hdfs.ext.dataframe import read_dataframe
     client = InsecureClient('http://nvwa01:50070')
-
-    df = read_dataframe(client,"/recommend/va/*.avro")
-
+    df = read_dataframe(client,"/recommend/native/part-00199-f83757ab-9f64-4a2c-9f27-0b76df51c1c4-c000.avro")
     print(df.head())
-    print(df.count())
+    # print(df.count())
 
     # spark.sql("use online")
     # spark.sql("ADD JAR /srv/apps/brickhouse-0.7.1-SNAPSHOT.jar")

tensnsorflow/record.py

@@ -28,6 +28,9 @@ def gen_tfrecords(in_file):
     from hdfs.ext.dataframe import read_dataframe
     client = InsecureClient('http://nvwa01:50070')
     df = read_dataframe(client,"/recommend/tr/part-00000-2f0d632b-0c61-4a0b-97d4-54bd5e579c5e-c000.avro")
+    df = df.rename({"app_list","level2_ids","level3_ids","stat_date","ucity_id", "ccity_name", "device_type", "manufacturer",
+                  "channel", "top", "time", "hospital_id","treatment_method", "price_min",
+                  "price_max", "treatment_time","maintain_time", "recover_time","y","z"})
 
     for i in range(df.shape[0]):
         feats = ["cid_id"]