修改cid、deviceid、pkl、model保存路径

diaryQueueUpdate.py

@@ -24,6 +24,7 @@ def get_video_id():
     db.close()
     return video_id
 
+
 # 将device_id、city_id拼接到对应的城市热门日记表。注意：下面预测集特征顺序要与训练集保持一致
 def feature_en(x_list, device_id):
     data = pd.DataFrame(x_list)

diaryTraining.py

@@ -10,11 +10,11 @@ def train():
     # log保存路径，如果不加这个参数，日志默认保存在/temp路径下，不符合规范
     param = {'task': 'binary', 'lr': lr, 'lambda': l2_lambda, 'metric': 'auc',"log":"/data2/models/result"}
 
-    ffm_model.fit(param, DIRECTORY_PATH + "model.out")
+    ffm_model.fit(param, DIRECTORY_PATH + "train/model.out")
 
     print("predicting")
     ffm_model.setTest(DIRECTORY_PATH + "test_ffm_data.csv")
     ffm_model.setSigmoid()
-    ffm_model.predict(DIRECTORY_PATH + "model.out",DIRECTORY_PATH + "test_set_predict_output.txt")
+    ffm_model.predict(DIRECTORY_PATH + "train/model.out",DIRECTORY_PATH + "test_set_predict_output.txt")
 
 

processData.py

@@ -52,14 +52,14 @@ def feature_en(data_start_date, data_end_date, validation_date, test_date):
     data_set_cid = data["cid"].unique()
     cid_df = pd.DataFrame()
     cid_df['cid'] = data_set_cid
-    cid_df.to_csv(DIRECTORY_PATH + "data_set_cid.csv", index=False)
+    cid_df.to_csv(DIRECTORY_PATH + "train/data_set_cid.csv", index=False)
     print("成功保存data_set_cid")
 
     # 将device_id 保存,目的是为了判断预测的device_id是否在这个集合里,如果不在，不需要预测
     data_set_device_id = data["device_id"].unique()
     device_id_df = pd.DataFrame()
     device_id_df['device_id'] = data_set_device_id
-    device_id_df.to_csv(DIRECTORY_PATH + "data_set_device_id.csv", index=False)
+    device_id_df.to_csv(DIRECTORY_PATH + "train/data_set_device_id.csv", index=False)
     print("成功保存data_set_device_id")
     return data, test_number, validation_number
 
@@ -69,7 +69,7 @@ def ffm_transform(data, test_number, validation_number):
     start = time.time()
     ffm_train = multiFFMFormatPandas()
     data = ffm_train.fit_transform(data, y='y',n=50000,processes=8)
-    with open(DIRECTORY_PATH+"ffm.pkl", "wb") as f:
+    with open(DIRECTORY_PATH+"train/ffm.pkl", "wb") as f:
         pickle.dump(ffm_train, f)
 
     print("done transform ffm")