LSSVR-python

LSSVR安装库函数需要python的版本大于等于3.8版本
LSSVR官网的网址：https://pypi.org/project/lssvr/

from lssvr import LSSVR
import numpy
import pandas as pd
from sklearn.model_selection import train_test_split
import joblib
import numpy as np
import matplotlib.pyplot as plt
from sklearn.metrics import mean_absolute_error
from sklearn.metrics import mean_squared_error

from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_squared_error
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import MinMaxScaler
from sklearn.metrics import mean_absolute_error

# 加载CSV文件
data = pd.read_csv('data.csv')
features = ['feature']
target = data['target ']
X_train, X_test, y_train, y_test = train_test_split(data[features], target, test_size=0.2, random_state=300)
# 特征归一化
scaler = MinMaxScaler()
X_train_scaled = scaler.fit_transform(X_train)
X_test_scaled = scaler.transform(X_test)

# 目标值归一化
target_scaler = MinMaxScaler()
y_train_scaled = target_scaler.fit_transform(y_train.values.reshape(-1, 1))
y_test_scaled = target_scaler.transform(y_test.values.reshape(-1, 1))


# 创建一个LSSVR模型，并设置相关参数
model = LSSVR(kernel='rbf', gamma=1.1)
# 拟合训练数据
model.fit(X_train_scaled, y_train_scaled.ravel())

# 在测试集上进行预测
y_pred_scaled = model.predict(X_test_scaled)

# 反向转换归一化的目标值
y_pred = target_scaler.inverse_transform(y_pred_scaled.reshape(-1, 1))

# 计算平均绝对误差（MAE）
mae = mean_absolute_error(y_test, y_pred)
print("Mean Absolute Error:", mae)
mse = mean_squared_error(y_test, y_pred)
print("Mean Squared Error:", mse)
from sklearn.metrics import r2_score

# 在训练集上进行预测
y_train_pred_scaled = model.predict(X_train_scaled)

# 反向转换归一化的目标值
y_train_pred = target_scaler.inverse_transform(y_train_pred_scaled.reshape(-1, 1))

# 计算R2
r2_train = r2_score(y_train, y_train_pred)
print("R2 train Score:", r2_train)
r2_test = r2_score(y_test, y_pred)
print("R2 test Score:", r2_test)

# 在训练集上进行预测
y_train_pred_scaled = model.predict(X_train_scaled)
y_train_pred = target_scaler.inverse_transform(y_train_pred_scaled.reshape(-1, 1))
fig = plt.figure(figsize=(15, 8), dpi=80)  # dpi越高放大越清楚
plt.rcParams['font.sans-serif'] = ['SimSun']  # 显示中文字体
plt.rcParams['axes.unicode_minus'] = False
# 绘制训练集预测值和真实值折线图
plt.subplot(121)
plt.plot(y_train_pred, 'ro-', label='训练集预测值')
plt.plot(np.asarray(y_train), 'bo-.', label='真实值')
plt.title("训练集")
plt.xlabel('样本序号')
plt.ylabel('值')
plt.legend(loc='best')

# 在测试集上进行预测
y_test_pred_scaled = model.predict(X_test_scaled)
y_test_pred = target_scaler.inverse_transform(y_test_pred_scaled.reshape(-1, 1))

# 绘制测试集预测值和真实值折线图
plt.subplot(122)
plt.plot(y_test_pred, 'ro-', label='测试集预测值')
plt.plot(np.asarray(y_test), 'bo-.', label='真实值')
plt.title("测试集")
plt.xlabel('样本序号')
plt.ylabel('值')
plt.legend(loc='best')
plt.show()