优化物业费评级中不合理

This commit is contained in:
shenjian@fangjia.com
2022-12-14 14:10:30 +08:00
parent cb771067df
commit 452334f320
@@ -1,13 +1,12 @@
"""
物业费等级计算
"""
from goojia_utils import ElasticSearchClient, nextcloudutils, dingutils, areautils
from goojia_utils import ElasticSearchClient, nextcloudutils
import pandas as pd
import logging
import os
from elasticsearch import helpers
import matplotlib.pyplot as plt
import numpy as np
from sklearn.mixture import GaussianMixture
from sklearn.preprocessing import QuantileTransformer
@@ -42,7 +41,7 @@ def gmm_train(total_rate, consumption_df, source, target):
# 高斯混合模型
city_info = consumption_df.loc[0, 'city']
if city_info in reasonable_dict and reasonable_dict[city_info] == '不合理':
gmm = GaussianMixture(n_components=total_rate, max_iter=200, n_init=5).fit(consumption_df[[source]])
gmm = GaussianMixture(n_components=total_rate, covariance_type='tied', max_iter=200, n_init=10).fit(consumption_df[[source]])
else:
gmm = GaussianMixture(n_components=total_rate).fit(consumption_df[[source]])
consumption_df['cluster'] = gmm.predict(consumption_df[[source]])
@@ -83,8 +82,10 @@ def calc_estate_charge_rating(community_summary_df):
count = 0
if len(df) < 5:
return []
df = gmm_train(5, df, source, target)
# 对于样本低于5的直接排名
df[target] = df['estate_charge'].rank(method='dense', ascending=False).astype(int)
else:
df = gmm_train(5, df, source, target)
# 批量更新
for index, item in df.iterrows():
row = dict()