心电图预测模型融合

模型融合

1. 简单加权融合:

- 回归（分类概率）：算术平均融合（Arithmetic mean），几何平均融合（Geometric mean）
取平均。稍稍改进的方法是进行加权平均。权值可以用排序的方法确定，
- 分类：投票（Voting)
假设对于一个二分类问题，有3个基础模型，那么就采取投票制的方法，投票多者确定为最终的分类。
- 综合：排序融合(Rank averaging)，log融合

2. stacking/blending:

- 构建多层模型，并利用预测结果再拟合预测。

Stacking 先从初始数据集训练出初级学习器，然后”生成”一个新数据集用于训练次级学习器。在这个新数据集中，初级学习器的输出被当作样例输入特征，而初始样本的标记仍被当作样例标记。
stacking一般使用交叉验证的方式，初始训练集D 被随机划分为k 个大小相似的集合D1 ， D2 ， … ， Dk，每次用k-1个部分训练T个模型，对另个一个部分产生T个预测值作为特征，遍历每一折后，也就得到了新的特征集合，标记还是源数据的标记，用新的特征集合训练一个集合模型。

Blending与Stacking大致相同，只是Blending的主要区别在于训练集不是通过K-Fold的CV策略来获得预测值从而生成第二阶段模型的特征，而是建立一个Holdout集，例如说10%的训练数据，第二阶段的stacker模型就基于第一阶段模型对这10%训练数据的预测值进行拟合。说白了，就是把Stacking流程中的K-Fold CV 改成 HoldOut CV

3. boosting/bagging（在xgboost，Adaboost,GBDT中已经用到）:

boosting

首先从训练集用初始权重训练出一个弱学习器1，根据弱学习的学习误差率表现来更新训练样本的权重，使得之前弱学习器1学习误差率高的训练样本点的权重变高，使得这些误差率高的点在后面的弱学习器2中得到更多的重视。然后基于调整权重后的训练集来训练弱学习器2；如此重复进行，直到弱学习器数达到事先指定的数目T，最终将这T个弱学习器通过集合策略进行整合，得到最终的强学习器。

bagging

采用有放回的方式进行抽样，用抽样的样本建立子模型,对子模型进行训练，这个过程重复多次，最后进行融合。

- 多树的提升方法
提升树
梯度提升树

import pandas as pd
import numpy as np
import warnings
import matplotlib
import matplotlib.pyplot as plt
import seaborn as sns

warnings.filterwarnings('ignore')
%matplotlib inline

import itertools
import matplotlib.gridspec as gridspec
from sklearn import datasets
from sklearn.linear_model import LogisticRegression
from sklearn.neighbors import KNeighborsClassifier
from sklearn.naive_bayes import GaussianNB 
from sklearn.ensemble import RandomForestClassifier,RandomForestRegressor
# from mlxtend.classifier import StackingClassifier
from sklearn.model_selection import cross_val_score, train_test_split
# from mlxtend.plotting import plot_learning_curves
# from mlxtend.plotting import plot_decision_regions

from sklearn.model_selection import StratifiedKFold
from sklearn.model_selection import train_test_split
from sklearn.model_selection import StratifiedKFold
from sklearn.model_selection import train_test_split
import lightgbm as lgb
from sklearn.neural_network import MLPClassifier,MLPRegressor
from sklearn.metrics import mean_squared_error, mean_absolute_error

def build_model_rf(X_train,y_train):
    model = RandomForestRegressor(n_estimators = 100)
    model.fit(X_train, y_train)
    return model


def build_model_lgb(X_train,y_train):
    model = lgb.LGBMRegressor(num_leaves=63,learning_rate = 0.1,n_estimators = 100)
    model.fit(X_train, y_train)
    return model


def build_model_nn(X_train,y_train):
    model = MLPRegressor(alpha=1e-05, hidden_layer_sizes=(5, 2), random_state=1,solver='lbfgs')
    model.fit(X_train, y_train)
    return model
    print('predict rf...')
model_rf = build_model_rf(X_train,y_train)
val_rf = model_rf.predict(X_val)
subA_rf = model_rf.predict(X_test)


print('predict lgb...')
model_lgb = build_model_lgb(X_train,y_train)
val_lgb = model_lgb.predict(X_val)
subA_lgb = model_rf.predict(X_test)


print('predict NN...')
model_nn = build_model_nn(X_train,y_train)
val_nn = model_nn.predict(X_val)
subA_nn = model_rf.predict(X_test)
    ```
    ```python
   
    def Weighted_method(test_pre1,test_pre2,test_pre3,w=[1/3,1/3,1/3]):
    Weighted_result = w[0]*pd.Series(test_pre1)+w[1]*pd.Series(test_pre2)+w[2]*pd.Series(test_pre3)
    return Weighted_result

# 初始权重，可以进行自定义，这里我们随便设置一个权重
w = [0.2, 0.3, 0.5]

val_pre = Weighted_method(val_rf,val_lgb,val_nn,w)
MAE_Weighted = mean_absolute_error(y_val,val_pre)
print('MAE of Weighted of val:',MAE_Weighted)
## 预测数据部分
subA = Weighted_method(subA_rf,subA_lgb,subA_nn,w)

## 生成提交文件
sub = pd.DataFrame()
sub['SaleID'] = X_test.index
sub['price'] = subA
sub.to_csv('./sub_Weighted.csv',index=False)

train_rf_pred = model_rf.predict(X_train)
train_lgb_pred = model_lgb.predict(X_train)
train_nn_pred = model_nn.predict(X_train)

stacking_X_train = pd.DataFrame()
stacking_X_train['Method_1'] = train_rf_pred
stacking_X_train['Method_2'] = train_lgb_pred
stacking_X_train['Method_3'] = train_nn_pred

stacking_X_val = pd.DataFrame()
stacking_X_val['Method_1'] = val_rf
stacking_X_val['Method_2'] = val_lgb
stacking_X_val['Method_3'] = val_nn

stacking_X_test = pd.DataFrame()
stacking_X_test['Method_1'] = subA_rf
stacking_X_test['Method_2'] = subA_lgb
stacking_X_test['Method_3'] = subA_nn
model_lr_stacking = build_model_rf(stacking_X_train,y_train)
train_pre_Stacking = model_lr_stacking.predict(stacking_X_train)
print('MAE of stacking:',mean_absolute_error(y_train,train_pre_Stacking))
val_pre_Stacking = model_lr_stacking.predict(stacking_X_val)
print('MAE of stacking:',mean_absolute_error(y_val,val_pre_Stacking))
print('Predict stacking...')
subA_Stacking = model_lr_stacking.predict(stacking_X_test)