Skip to content

M
model_mvp
Commit 085d706c authored by 王家华's avatar 王家华
Browse files
Options

仍然没有加表格 好气呀

parent 9b10189a
Expand all Hide whitespace changes
Inline Side-by-side
Showing with 130 additions and 99 deletions
.idea/workspace.xml
View file @ 085d706c
This diff is collapsed.
graph/matplot.py
View file @ 085d706c
......@@ -57,7 +57,7 @@ def plot_table_list(datalist, auc, datalist_description=None, title='untitled',
# 每个table需要只有一个index，一个values
x = range(len(datalist[table_index].index))
y = datalist[table_index].values
axs.plot(x, y, label=datalist_description[table_index])
axs.plot(x, y, label=datalist_description[table_index] + "AUC: "+ str(round(auc[table_index],3)))
if len(x) == 1:
plot_tab = False
if plot_tab:
......@@ -100,7 +100,7 @@ def plot_table_list(datalist, auc, datalist_description=None, title='untitled',
plt.xticks([])
# otherwise, nothing to do here
the_table.auto_set_font_size(False)
the_table.set_fontsize(8)
the_table.set_fontsize(6)
fig.subplots_adjust(bottom=0.2)
plt.grid()
if y_label is not None:
......
lgb_report.docx 0 → 100644
View file @ 085d706c
File added
models_kit/general_methods.py
View file @ 085d706c
......@@ -39,24 +39,26 @@ def topN_feature_importance_list(features, clf, topN=3):
return importanct_feat
def model_selection(algorthm,clf,df_train,df_val,df_test,target,score,optimal_model,model_obj):
# model matrix 存储不同模型指标的矩阵
model_matrix_index = ['name', 'Params', 'trainAUC', 'validationAUC']
model_matrix = pd.DataFrame(['NULL', 'NULL', roc_auc_score(df_train[target], df_train[score]),
roc_auc_score(df_train[target], df_train[score])], index=model_matrix_index,
columns=['线上模型'])
#TODO here
# 定义最优参指针
pointer = 0
# 遍历最优参组合
for param in optimal_para:
if algorthm == "lightGBM":
train_auc, val_auc, lgbm = lightgbm.train_lgbm(lightgbm.params_lgb, df_train, df_val, model_obj.features,
adds_on=param, target=target)
model_matrix = pd.concat([model_matrix,
pd.DataFrame(['lightGBM', param, train_auc, val_auc], index=model_matrix_index,
columns=[pointer])], axis=1)
pointer += 1
# 简单选取一下validation set auc 最高的 params
best_params = model_matrix.T.sort_values(by='validationAUC', ascending=False).iloc[0, :].loc['Params']
\ No newline at end of file
# def model_selection(algorthm,clf,df_train,df_val,df_test,target,score,optimal_model,model_obj):
# # model matrix 存储不同模型指标的矩阵
# model_matrix_index = ['name', 'Params', 'trainAUC', 'validationAUC']
# model_matrix = pd.DataFrame(['NULL', 'NULL', roc_auc_score(df_train[target], df_train[score]),
# roc_auc_score(df_train[target], df_train[score])], index=model_matrix_index,
# columns=['线上模型'])
#
# # 定义最优参指针
# pointer = 0
# # 遍历最优参组合
# for param in optimal_para:
# if algorthm == "lightGBM":
# train_auc, val_auc, lgbm = lightgbm.train_lgbm(lightgbm.params_lgb, df_train, df_val, model_obj.features,
# adds_on=param, target=target)
# model_matrix = pd.concat([model_matrix,
# pd.DataFrame(['lightGBM', param, train_auc, val_auc], index=model_matrix_index,
# columns=[pointer])], axis=1)
# pointer += 1
#
# # 简单选取一下validation set auc 最高的 params
# best_params = model_matrix.T.sort_values(by='validationAUC', ascending=False).iloc[0, :].loc['Params']
\ No newline at end of file
mvp/allocator.py
View file @ 085d706c
......@@ -62,11 +62,11 @@ df_train, df_val, df_test = datacal.train_test_split_general(df_sample, val_size
del df_sample
# 模型refit
model_matrix, lgbm = refit.model_fit(df_sample, dhb, target, score)
model_matrix, lgbm = refit.model_fit(df_train, df_val, df_test, dhb, target, score)
print(model_matrix)
# 生成报告
status = refit.model_report(lgbm, df_train, df_val, df_test, dhb, target,
status = refit.model_report(lgbm, df_train, df_val, df_test, dhb, target, model_matrix,
score, prediction, report_path, report_name, applied_from, applied_type, topN=3)
......
mvp/refit.py
View file @ 085d706c
......@@ -83,6 +83,9 @@ def model_report(clf, df_train, df_val, df_test, model_obj, target,model_matrix,
:return:
status ： 返回1表示执行完成
'''
# 样本分布
df_train_ = len(df_train)
# 用新模型预测结果 xgb还需要加一个proba (TODO here)
predictions ,test_auc = lightgbm.predict(clf,df_test,model_obj.features,target)
......@@ -126,7 +129,7 @@ def model_report(clf, df_train, df_val, df_test, model_obj, target,model_matrix,
#univarChart = matplot.plot_table_list([univar_train,univar_val,univar_test], [1,2,3], datalist_description=None, title= i +' univar Chart', X_label=None, y_label=None,
# tab_df_list=tab_df_list, plot_tab=False,
# saved_path='./mvp/plots/cache/')
document.add_picture(univar_chart,width=Inches(8))
document.add_picture(univar_chart,width=Inches(7))
# 新增pdp段
document.add_paragraph('PDP_chart')
......@@ -136,45 +139,65 @@ def model_report(clf, df_train, df_val, df_test, model_obj, target,model_matrix,
# pdpChart = matplot.plot_table_df(pdp, ['1'], title=i + ' PDP Chart', X_label=None, y_label=None,
# tab_df=None, plot_tab=True, saved_path='./mvp/plots/cache/')
pdpChart = matplot.pdpCharts9(clf, df_test, importanct_feat, model_obj.features, n_bins=10, dfltValue=-99999, maxValRatio=1, saved_path="./mvp/plots/cache/")
document.add_picture(pdpChart,width=Inches(8))
document.add_picture(pdpChart,width=Inches(7))
# 新增liftchart段
document.add_paragraph('lift_chart')
# 遍历给定渠道 & 客群 默认等频画出liftchart
try:
lift_pred = datacal.cal_lift(df_test, score=prediction)
lift_online = datacal.cal_lift(df_test, score=score)
# liftChart = matplot.plot_table(lift, title=i +' lift Chart',saved_path='./mvp/plots/cache/')
liftChart = matplot.plot_table_list([lift_pred['mean'], lift_online['mean']],
[roc_auc_score(df_test[target], df_test[prediction]),
roc_auc_score(df_test[target], df_test[score])], datalist_description=None,
roc_auc_score(df_test[target], df_test[score])], datalist_description=['新模型预测','线上模型'],
title='全渠道全量客群测试集上的 lift Chart',
X_label=None, y_label=None,
tab_df_list=None, plot_tab=False,
X_label=None, y_label='逾期率',
tab_df_list=[lift_pred['count'], lift_online['count']], plot_tab=False,
saved_path='./mvp/plots/cache/')
document.add_picture(liftChart, width=Inches(8))
# 遍历渠道
document.add_picture(liftChart, width=Inches(7))
# 遍历渠道
for channel in applied_from.keys():
# 遍历客群类型
for type in applied_type.keys():
print('lift ',type,channel)
print('lift ',channel)
# 数据切片
df_sliced = df_test[
df_test.applied_type.map(lambda x: True if str(x) in type.split(',') else False) & df_test.applied_from.map(
lambda x: True if str(x) in channel.split(',') else False)]
#
lift_pred = datacal.cal_lift(df_sliced, score=prediction)
lift_online = datacal.cal_lift(df_sliced, score=score)
# liftChart = matplot.plot_table(lift, title=i +' lift Chart',saved_path='./mvp/plots/cache/')
liftChart = matplot.plot_table_list([lift_pred['mean'], lift_online['mean']],
[roc_auc_score(df_test[target], df_test[prediction]),
roc_auc_score(df_test[target], df_test[score])],
datalist_description=['新模型预测', '线上模型'],
title=applied_from[channel] + ' lift Chart',
X_label=None, y_label='逾期率',
tab_df_list=[lift_pred['count'], lift_online['count']], plot_tab=False,
saved_path='./mvp/plots/cache/')
document.add_picture(liftChart, width=Inches(5.5))
# 遍历客群类型
for type in applied_type.keys():
print('lift ',type)
# 数据切片
df_sliced = df_test[df_test.applied_type.map(lambda x : True if str(x) in type.split(',') else False) & df_test.applied_from.map(lambda x : True if str(x) in channel.split(',') else False)]
#
lift_pred = datacal.cal_lift(df_sliced,score=prediction)
lift_online = datacal.cal_lift(df_sliced,score=score)
#liftChart = matplot.plot_table(lift, title=i +' lift Chart',saved_path='./mvp/plots/cache/')
liftChart = matplot.plot_table_list([lift_pred['mean'], lift_online['mean']], [roc_auc_score(df_test[target],df_test[prediction]),roc_auc_score(df_test[target],df_test[score])], datalist_description=None,
title= applied_from[channel]+applied_type[type]+ ' lift Chart', X_label=None, y_label=None,
tab_df_list=[lift_pred['count'], lift_online['count']], plot_tab=True,
liftChart = matplot.plot_table_list([lift_pred['mean'], lift_online['mean']], [roc_auc_score(df_test[target],df_test[prediction]),roc_auc_score(df_test[target],df_test[score])], datalist_description=['新模型预测','线上模型'],
title= applied_type[type]+ ' lift Chart', X_label=None, y_label='逾期率',
tab_df_list=[lift_pred['count'], lift_online['count']], plot_tab=False,
saved_path='./mvp/plots/cache/')
document.add_picture(liftChart,width=Inches(8))
# 存在某些渠道量很少的情况，加入try catch异常处理
document.add_picture(liftChart,width=Inches(5.5))
except Exception as e:
print('Exception: ',e)
print(e)
pass
# 存在某些渠道量很少的情况，加入try catch异常处理
# docx 保存
filetool.saveDocument(document, report_path, report_name)
return 1
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment