（笔记）
超参数：在算法运行前需要决定的参数
模型参数：算法过程中学习的参数
import numpy as np
from sklearn import datasets
from sklearn.model_selection import train_test_split # 拆分训练数据，测试数据
from sklearn.neighbors import KNeighborsClassifier #knn分类器
digits = datasets.load_digits()
X = digits.data
y = digits.target
X_train,X_test,y_train,y_test = train_test_split(X, y, test_size=0.2, random_state=666)
knn_clf = KNeighborsClassifier(n_neighbors=3)
knn_clf.fit(X_train,y_train)
predict = knn_clf.score(X_test,y_test)
print(predict) # 0.9888888888888889
输出结果：0.9888888888888889
这是一个最简单的算法，输出的结果是预测的准确率（就是y_train的预测数据和y_test比较）
但是此时我们有一个疑问 为什么n_neighbors=3 这是我们默认写的，就是最好的吗？我们如何才能求到最好的k值？
import numpy as np
from sklearn import datasets
from sklearn.model_selection import train_test_split
from sklearn.neighbors import KNeighborsClassifier
digits = datasets.load_digits()
X = digits.data
y = digits.target
X_train,X_test,y_train,y_test = train_test_split(X, y, test_size=0.2, random_state=666)
#寻找最好的k
best_score = 0.0
best_k = -1
for k in range(1, 11): #意思取最近的1-10个数进行比较
knn_clf = KNeighborsClassifier(n_neighbors=k)
knn_clf.fit(X_train, y_train)
score = knn_clf.score(X_test, y_test)
if score > best_score:
best_k = k
best_score = score
print("best_k =", best_k) # best_k = 4
print("best_score =", best_score) # best_score = 0.9916666666666667
注意：如果计算出的best_k是在边界，比如是10，那么我们需要在往外延伸，因为可能10是这之间最好的，但不是真正最好的，最好的可能k可能比10大，这是我们可以继续range(8-21)，依次知道best_k不在是边界为止
到目前为止我们只考虑了k值，但其实k近邻是比较离xxx最近的K个样本，那么我们是不是应该也要把距离的因素考虑进去？（目前我们的距离都是指欧拉距离）
更新代码如下
import numpy as np
from sklearn import datasets
from sklearn.model_selection import train_test_split # 拆分训练数据，测试数据
from sklearn.neighbors import KNeighborsClassifier #knn分类器
digits = datasets.load_digits()
X = digits.data
y = digits.target
X_train,X_test,y_train,y_test = train_test_split(X, y, test_size=0.2, random_state=666)
best_score = 0.0
best_k = -1
best_method = ""
for method in ["uniform", "distance"]: #因为KNeighborsClassifier有参数weights表示是否需要考虑距离，默认uniform不考虑
for k in range(1, 11):
knn_clf = KNeighborsClassifier(n_neighbors=k, weights=method)
knn_clf.fit(X_train, y_train)
score = knn_clf.score(X_test, y_test)
if score > best_score:
best_k = k
best_score = score
best_method = method
print("best_method =", best_method)
print("best_k =", best_k)
print("best_score =", best_score)
# best_method = uniform
# best_k = 4
# best_score = 0.9916666666666667
但是考虑到距离的问题，我们还会考虑到何为距离
我们目前用的是欧拉距离，如果使用明可夫斯基距离（(Minkowski Distance）呢？明可夫斯基是从欧拉，曼哈顿推导出来的公式（p=1时明可夫斯基距离相当于曼哈顿距离，p=2时明可夫斯基距离相当于欧拉距离）
具体过程看https://blog.csdn.net/xiaoduan_/article/details/79327781
所以这是我们会想到还有另一个超参数p
import numpy as np
from sklearn import datasets
from sklearn.model_selection import train_test_split # 拆分训练数据，测试数据
from sklearn.neighbors import KNeighborsClassifier #knn分类器
digits = datasets.load_digits()
X = digits.data
y = digits.target
X_train,X_test,y_train,y_test = train_test_split(X, y, test_size=0.2, random_state=666)
best_score = 0.0
best_k = -1
best_p= -1
for k in range(1, 11):
for p in range(1, 5):
knn_clf = KNeighborsClassifier(n_neighbors=k, weights="distance",p=p)
knn_clf.fit(X_train, y_train)
score = knn_clf.score(X_test, y_test)
if score > best_score:
best_k = k
best_score = score
best_p = p
print("best_p =", p)
print("best_k =", best_k)
print("best_score =", best_score)
# best_p = 4
# best_k = 3
# best_score = 0.9888888888888889
上述代码的网格搜索，（两层循坏k*p的网格）存在问题是寻找best_p，的时候默认了weights=distance，所以best_method和best_p在上述代码无法一起处理
修改如下：
import numpy as np
from sklearn import datasets
from sklearn.model_selection import train_test_split # 拆分训练数据，测试数据
from sklearn.neighbors import KNeighborsClassifier #knn分类器
from sklearn.model_selection import GridSearchCV
digits = datasets.load_digits()
X = digits.data
y = digits.target
X_train,X_test,y_train,y_test = train_test_split(X, y, test_size=0.2, random_state=666)
param_grid = [
{
'weights': ['uniform'],
'n_neighbors': [i for i in range(1, 11)]
},
{
'weights': ['distance'],
'n_neighbors': [i for i in range(1, 11)],
'p': [i for i in range(1, 6)]
}
]
knn_clf = KNeighborsClassifier()
grid_search = GridSearchCV(knn_clf, param_grid, n_jobs=-1, verbose=2) # n_jobs是使用多少个cpu，-1表示所有，verbose表示打印输出信息
grid_search.fit(X_train, y_train)
print(grid_search.best_score_)
因为sklearn已经为我们封装好了网格搜索的方法 所以直接导入：from sklearn.model_selection import GridSearchCV
然后将自己的网格信息存入变量param_grid ，最后调用GridSearchCV方法并打印结果
