优化财力评级

This commit is contained in:
沈健
2025-04-11 15:06:50 +08:00
parent 608f492b29
commit ddfb1cde70
2 changed files with 18 additions and 44 deletions
+17 -43
View File
@@ -12,6 +12,7 @@ from elasticsearch import helpers
import matplotlib.pyplot as plt
import sys
from sklearn.preprocessing import RobustScaler
from sklearn.mixture import GaussianMixture
from sklearn.preprocessing import QuantileTransformer
@@ -31,6 +32,7 @@ nextcloud_aim_file_path = '共享/财力评级/财力评级图'
reasonable_df = pd.read_excel(path + '/财力评级分布情况20221215.xlsx')
reasonable_dict = reasonable_df.set_index('city')['是否合理'].to_dict()
scaler = RobustScaler()
# Min-Max 归一化到 [1, 100]
@@ -72,14 +74,17 @@ def calc_consumption_potential(city):
# 户均总价计算时候改成万
community_summary_df['total_price'] = round(community_summary_df['total_price'] / 10000, 0)
# 噪音词处理
community_summary_df['total_price_fixed'] = QuantileTransformer(output_distribution="normal").fit_transform(
community_summary_df[['total_price']])
community_summary_df['avg_cost_fixed'] = QuantileTransformer(output_distribution="normal").fit_transform(
community_summary_df[['avg_cost']])
community_summary_df['total_price_fixed'] = community_summary_df['total_price']
community_summary_df['avg_cost_fixed'] = community_summary_df['avg_cost']
community_summary_df['estate_charge_fixed'] = QuantileTransformer(output_distribution="normal").fit_transform(
community_summary_df[['estate_charge_fix']])
community_summary_df['estate_charge_fixed'] = community_summary_df['estate_charge_fix']
# community_summary_df['total_price_fixed'] = QuantileTransformer(output_distribution="normal").fit_transform(
# community_summary_df[['total_price']])
# community_summary_df['avg_cost_fixed'] = QuantileTransformer(output_distribution="normal").fit_transform(
# community_summary_df[['avg_cost']])
#
# community_summary_df['estate_charge_fixed'] = QuantileTransformer(output_distribution="normal").fit_transform(
# community_summary_df[['estate_charge_fix']])
community_summary_df['estate_charge_mean'] = community_summary_df.groupby(by=['city'])['estate_charge_fixed'].transform(
np.mean)
@@ -166,7 +171,8 @@ def calc_consumption_rating(community_summary_df):
else:
# 极值设置
score_df = gmm_train(total_rate, df.dropna(subset=['total_price_score']), "total_price_score", "total_price_rating")
score_df = min_max_scale(score_df, 'total_price_score', 'total_price_score_minmax')
score_df['robust_total_price_score'] = scaler.fit_transform(score_df[['total_price_score']])
score_df = min_max_scale(score_df, 'robust_total_price_score', 'total_price_score_minmax')
df = filter_min_max(df)
if len(df.dropna(subset=['consumption_potential'])) >= total_rate:
df = gmm_train(total_rate, df.dropna(subset=['consumption_potential']), source, target)
@@ -176,7 +182,9 @@ def calc_consumption_rating(community_summary_df):
df = gmm_train(total_rate, df, "avg_cost_score", "avg_cost_rating")
# 应用归一化
df = min_max_scale(df, 'consumption_potential', 'consumption_score')
df['robust_potential'] = scaler.fit_transform(df[['consumption_potential']])
df = min_max_scale(df, 'robust_potential', 'consumption_score') # 再映射到 [1, 100]
#df = min_max_scale(df, 'consumption_potential', 'consumption_score')
if len(score_df) > 0:
# 补充财力评级
df = pd.merge(score_df[['_id', 'total_price_rating', 'total_price_score_minmax']], df, how='left', on=['_id'])
@@ -289,40 +297,6 @@ def sync(city=None):
def filter_min_max(community_summary_df):
community_summary_df['normal_consumption_potential'] = QuantileTransformer(output_distribution="normal").fit_transform(community_summary_df[['consumption_potential']])
# total_price5 = np.percentile(a=community_summary_df['total_price'], q=2, axis=0)
# total_price95 = np.percentile(a=community_summary_df['total_price'], q=98, axis=0)
# community_summary_df.loc[community_summary_df['total_price'] > total_price95, 'total_price'] = total_price95
# community_summary_df.loc[community_summary_df['total_price'] < total_price5, 'total_price'] = total_price5
#
# estate_charge5 = np.percentile(a=community_summary_df['estate_charge'], q=2, axis=0)
# estate_charge95 = np.percentile(a=community_summary_df['estate_charge'], q=98, axis=0)
# community_summary_df.loc[community_summary_df['estate_charge'] > estate_charge95, 'estate_charge'] = estate_charge95
# community_summary_df.loc[community_summary_df['estate_charge'] < estate_charge5, 'estate_charge'] = estate_charge5
#
# avg_cost5 = np.percentile(a=community_summary_df['avg_cost'], q=2, axis=0)
# avg_cost95 = np.percentile(a=community_summary_df['avg_cost'], q=98, axis=0)
# community_summary_df.loc[community_summary_df['avg_cost'] > avg_cost95, 'avg_cost'] = avg_cost95
# community_summary_df.loc[community_summary_df['avg_cost'] < avg_cost5, 'avg_cost'] = avg_cost5
#if city in factor_dict:
# 户均总价按极值设置可靠性不高,当户均总价补充后,图中大于最大值的量非常多
# community_summary_df.loc[
# community_summary_df['total_price'] < factor_dict[city]['min_total_price'], 'total_price'] = \
# factor_dict[city]['min_total_price']
# community_summary_df.loc[
# community_summary_df['total_price'] > factor_dict[city]['max_total_price'], 'total_price'] = \
# factor_dict[city]['max_total_price']
# community_summary_df.loc[
# community_summary_df['estate_charge'] < factor_dict[city]['min_estate_charge'], 'estate_charge'] = \
# factor_dict[city]['min_estate_charge']
# community_summary_df.loc[
# community_summary_df['estate_charge'] > factor_dict[city]['max_estate_charge'], 'estate_charge'] = \
# factor_dict[city]['max_estate_charge']
# community_summary_df.loc[community_summary_df['avg_cost'] < factor_dict[city]['min_avg_cost'], 'avg_cost'] = \
# factor_dict[city]['min_avg_cost']
# community_summary_df.loc[community_summary_df['avg_cost'] > factor_dict[city]['max_avg_cost'], 'avg_cost'] = \
# factor_dict[city]['max_avg_cost']
return community_summary_df
@@ -146,7 +146,7 @@ def detail(city):
'周边',
'自有'
),
np.nan # 如果 estate_charge 为空,则返回 NaN
''
)
community_summary_df['credit'] = np.where(community_summary_df['total_price_diff'] == 0, '高', np.where(community_summary_df['total_price_diff'] == 1, '中', np.where(community_summary_df['total_price_diff'] >= 2, '低', '-')))