kaggle Partial_Dependence_Plots

# Partial dependence plots
# 改变单变量对最终预测结果的影响
# 先fit出一种模型，然后取一行，不断改变某一特征，看它对最终结果的印象。
# 但是，只使用一行不具有典型性
# 所以对所有行执行上述操作，求均值

 

import pandas as pd
from sklearn.ensemble import GradientBoostingRegressor, GradientBoostingClassifier
from sklearn.ensemble.partial_dependence import partial_dependence, plot_partial_dependence
from sklearn.preprocessing import Imputer
import matplotlib.pyplot as plt 

train_path = r"C:\Users\cbattle\Desktop\train.csv"
test_path = r"C:\Users\cbattle\Desktop\test.csv"
out_path = r"C:\Users\cbattle\Desktop\out.csv"

def get_some_data():
    data = pd.read_csv(train_path)
    y = data.SalePrice
    cols_to_use = ['YearBuilt', 'GrLivArea', 'TotRmsAbvGrd']
    X = data[cols_to_use]
    my_imputer = Imputer()
    imputed_X = my_imputer.fit_transform(X)
    return imputed_X, y
    
X, y = get_some_data()
my_model = GradientBoostingRegressor()
my_model.fit(X, y)
my_plots = plot_partial_dependence(my_model, 
                                   features=[0,2], 
                                   X=X, 
                                   feature_names=cols_to_use, 
                                   grid_resolution=10)
plt.show()
# print('ok')

# There is a function called partial_dependence to get the raw data making up this plot, rather than making the visual plot itself.
# This is useful if you want to control how it is visualized using a plotting package like Seaborn. With moderate effort, you could
# make much nicer looking plots.