scikit-Learn demo

一、大致流程

1.Estimator框架（预测器）

• model.fit()

2.Supervised Learning

• model.predict()
• model.predict_proba()
• model.score()

3.Unsupervised learning

* 没有标注型数据，不知道最终预测的结果，需要对数据进行转换

• model.transform()
• model.fit_transform() #效率更高

1.Practice regression

以Boston房价为例

a.LinearRegression

%matplotlib inline
import matplotlib.pyplot as plt
from sklearn.datasets import load_boston
from sklearn.linear_model import LinearRegression
from sklearn.metrics import mean_absolute_error
# 载入数据
data = load_boston()
# 样本数，样本特征数
n_samples, n_feature = data.data.shape
# new 一个lr模型
lr = LinearRegression()
# 用模型拟合数据
lr.fit(data.data, data.target)
# 预测数据
pred = lr.predict(data.data)
# 评估模型
score = mean_absolute_error(data.target, pred)
print(score,n_samples,n_feature)
print(data.target.shape, data.keys())

3.272944637996926 506 13
(506,) dict_keys([‘data’, ‘target’, ‘feature_names’, ‘DESCR’])

plt.scatter(data.target, pred)
plt.xlabel('true price')
plt.ylabel('predicted price')
plt.plot(data.target, data.target, c='r')

# play with the feature (feature_i -- target) 
column_i = 5
plt.scatter(data.data[:,column_i], data.target)
print(data.feature_names[column_i])

b.DecisionTree

from sklearn.tree import DecisionTreeRegressor
dtr = DecisionTreeRegressor()
dtr.fit(data.data, data.target)
pred_dtr = dtr.predict(data.data