service model 优化

test/model_test.py

@@ -164,18 +164,33 @@ def test(df_train):
     FEATURE_COLUMN_KEY = "Strategy:rec:column:service:" + VERSION
     conn =  connUtils.getRedisConn4()
     s = conn.get(FEATURE_COLUMN_KEY)
-    # s = conn.get(FEATURE_USER_KEY+"869982038583034")
+    user = conn.get(FEATURE_USER_KEY+"869982038583034")
+    item = conn.get(FEATURE_ITEM_KEY+"5884526")
     datasColumnss = json.loads(s,encoding='utf-8')
+    user_d = json.loads(user,encoding='utf-8')
+    item_d = json.loads(item,encoding='utf-8')
+
+    ddd = {}
+    for d in datasColumnss:
+        res = ["-1"]
+        if d in user_d:
+            res = [user_d[d]]
+        if d in item_d:
+            res = [item_d[d]]
+
+        ddd[d] = res
+
+
     #
     # conn.get()
 
 
-    ddd = {}
-    datasColumnss = df_train.columns.to_list()
-    dd = df_train.sample(n=10)
-    for c in datasColumnss:
-        vvv = dd[c].tolist()
-        ddd[c] = vvv
+    # ddd = {}
+    # datasColumnss = df_train.columns.to_list()
+    # dd = df_train.sample(n=10)
+    # for c in datasColumnss:
+    #     vvv = dd[c].tolist()
+    #     ddd[c] = vvv
 
     pre_data = {"inputs":ddd}
     pre_data = json.dumps(pre_data)
@@ -187,49 +202,49 @@ def test(df_train):
 
 if __name__ == '__main__':
 
-    curTime = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
-    print("train_service执行时间:{}".format(curTime))
-    splitTimestamp = int(time.mktime(time.strptime(addDays(-1), "%Y%m%d")))
-
-    # redis中加载数据字典
-    print("加载模型字典...")
-    data_vocab = json.load(open(configUtils.VOCAB_PATH,mode='r',encoding='utf-8'))
-    print("字典keys:",str(data_vocab.keys()))
-    # data_vocab = getDataVocabFromRedis(VERSION)
-    assert data_vocab
+    # curTime = time.strftime("%Y-%m-%d %H:%M:%S", time.localtime())
+    # print("train_service执行时间:{}".format(curTime))
+    # splitTimestamp = int(time.mktime(time.strptime(addDays(-1), "%Y%m%d")))
+    #
+    # # redis中加载数据字典
+    # print("加载模型字典...")
+    # data_vocab = json.load(open(configUtils.VOCAB_PATH,mode='r',encoding='utf-8'))
+    # print("字典keys:",str(data_vocab.keys()))
+    # # data_vocab = getDataVocabFromRedis(VERSION)
+    # assert data_vocab
+    #
+    # timestmp1 = int(round(time.time()))
+    # df_train = loadData(data_path_train)
+    # print(df_train.dtypes)
+    # print("训练数据列：",df_train.columns)
+    #
+    # df_test = df_train.loc[df_train['timestamp']>=splitTimestamp]
+    # df_train = df_train.loc[df_train['timestamp'] < splitTimestamp]
+    # # df_test = loadData(data_path_test)
+    # timestmp2 = int(round(time.time()))
+    # print("读取数据耗时s:{}".format(timestmp2 - timestmp1))
+    #
+    # # 获取训练列
+    # columns = df_train.columns.tolist()
+    # print("原始数据列：")
+    # print(columns)
+    # emb_columns,number_columns,inputs = getTrainColumns(columns, data_vocab)
+    # print("训练列：")
+    # datasColumns = list(inputs.keys())
+    # print(datasColumns)
+    #
+    # df_train = df_train[datasColumns + ["label"]]
+    # df_test = df_test[datasColumns + ["label"]]
+    #
+    # trainSize = df_train["label"].count()
+    # print("trainSize:{}".format(trainSize))
+    # testSize = df_test["label"].count()
+    # print("trainSize:{},testSize{}".format(trainSize,testSize))
+    #
+    # # 数据类型转换
+    # df_train = csvTypeConvert(datasColumns,df_train,data_vocab)
+    # df_test = csvTypeConvert(datasColumns,df_test,data_vocab)
 
-    timestmp1 = int(round(time.time()))
-    df_train = loadData(data_path_train)
-    print(df_train.dtypes)
-    print("训练数据列：",df_train.columns)
-
-    df_test = df_train.loc[df_train['timestamp']>=splitTimestamp]
-    df_train = df_train.loc[df_train['timestamp'] < splitTimestamp]
-    # df_test = loadData(data_path_test)
-    timestmp2 = int(round(time.time()))
-    print("读取数据耗时s:{}".format(timestmp2 - timestmp1))
-
-    # 获取训练列
-    columns = df_train.columns.tolist()
-    print("原始数据列：")
-    print(columns)
-    emb_columns,number_columns,inputs = getTrainColumns(columns, data_vocab)
-    print("训练列：")
-    datasColumns = list(inputs.keys())
-    print(datasColumns)
-
-    df_train = df_train[datasColumns + ["label"]]
-    df_test = df_test[datasColumns + ["label"]]
-
-    trainSize = df_train["label"].count()
-    print("trainSize:{}".format(trainSize))
-    testSize = df_test["label"].count()
-    print("trainSize:{},testSize{}".format(trainSize,testSize))
-
-    # 数据类型转换
-    df_train = csvTypeConvert(datasColumns,df_train,data_vocab)
-    df_test = csvTypeConvert(datasColumns,df_test,data_vocab)
-
-    test(df_train)
+    test(None)