|
|
@@ -53,7 +53,7 @@ def get_data_from_odps(project, table, num):
|
|
|
if reader.count < num:
|
|
|
return None
|
|
|
# 获取字段名称
|
|
|
- column_names = reader.schema.names
|
|
|
+ column_names = [col.name for col in reader.get_schema().columns]
|
|
|
# 获取查询结果数据
|
|
|
data = []
|
|
|
for record in reader:
|
|
|
@@ -564,6 +564,7 @@ def validate_model_data_accuracy(start_time):
|
|
|
# 对指定列取绝对值再求和
|
|
|
old_abs_avg = df['old_error'].abs().sum() / num
|
|
|
new_abs_avg = df['new_error'].abs().sum() / num
|
|
|
+ print(old_abs_avg, new_abs_avg)
|
|
|
new_auc = get_dict_from_odps('pai_algo', table_dict['二分类评估-1'])['AUC']
|
|
|
old_auc = get_dict_from_odps('pai_algo', table_dict['二分类评估-2'])['AUC']
|
|
|
bizdate = get_previous_days_date(1)
|
|
|
@@ -598,17 +599,19 @@ def validate_model_data_accuracy(start_time):
|
|
|
msg += f"\n\t - 新模型AUC: {new_auc}"
|
|
|
msg += f"\n\t - 老模型Top10差异平均值: {old_abs_avg}"
|
|
|
msg += f"\n\t - 新模型Top10差异平均值: {new_abs_avg}"
|
|
|
- _monitor(level, msg, start_time, elapsed, top10_msg)
|
|
|
+ print(level, msg, start_time, elapsed, top10_msg)
|
|
|
|
|
|
|
|
|
if __name__ == '__main__':
|
|
|
start_time = int(time.time())
|
|
|
- # 1.更新工作流
|
|
|
- update_online_flow()
|
|
|
- # 2.训练模型
|
|
|
- train_res = train_model()
|
|
|
- if train_res:
|
|
|
- # 3. 验证模型数据 & 更新模型到线上
|
|
|
- validate_model_data_accuracy(start_time)
|
|
|
- else:
|
|
|
- print('train_model_error')
|
|
|
+ validate_model_data_accuracy(start_time)
|
|
|
+ # start_time = int(time.time())
|
|
|
+ # # 1.更新工作流
|
|
|
+ # update_online_flow()
|
|
|
+ # # 2.训练模型
|
|
|
+ # train_res = train_model()
|
|
|
+ # if train_res:
|
|
|
+ # # 3. 验证模型数据 & 更新模型到线上
|
|
|
+ # validate_model_data_accuracy(start_time)
|
|
|
+ # else:
|
|
|
+ # print('train_model_error')
|
xueyiming