机器学习-自定义Loss函数

1、简介

机器学习框架中使用自定义的Loss函数，

2、应用

（1）sklearn

python 复制代码

from sklearn.metrics import max_error
from sklearn.metrics import make_scorer
from sklearn.model_selection import cross_val_score
from sklearn.linear_model import Ridge

def custom_loss(y_true, y_pred, **kwargs):
    # Define your custom loss calculation here
    y_true = np.array(y_true)
    y_pred = np.array(y_pred)
    if y_true.ndim == 1 :
        y_true = y_true.reshape((-1, 1))

    if y_pred.ndim == 1:
        y_pred = y_pred.reshape((-1, 1))   
        
    loss = max(y_true-y_pred)
    return loss


data = pd.DataFrame(np.array([[i for i in range(0,300)],[i for i in range(100,400)],[i for i in range(200,500)]]).T,columns=['a','b','c'])

X_train ,y_train = data[['a','b']],data[['c']]
clf = Ridge()

custom_scorer = make_scorer(custom_loss, greater_is_better=False)

# Create and train a model using the custom loss function
# model = Ridge()
scores = cross_val_score(clf, X_train, y_train, cv=5, scoring=custom_scorer)

输出是cv=5，交叉验证的5个结果，评估模型

（2）pycaret

python 复制代码

from pycaret.regression import *
from pycaret.datasets import get_data
import pandas as pd
import numpy as np
from sklearn.metrics import max_error
from sklearn.metrics import make_scorer


def custom_loss(y_true, y_pred, **kwargs):
    # Define your custom loss calculation here
    y_true = np.array(y_true)
    y_pred = np.array(y_pred)
    if y_true.ndim == 1 :
        y_true = y_true.reshape((-1, 1))

    if y_pred.ndim == 1:
        y_pred = y_pred.reshape((-1, 1))   
        
    loss = max(y_true-y_pred)
    return loss

# # load sample dataset
# # data = get_data('insurance')
data = pd.DataFrame(np.array([[i for i in range(0,300)],[i for i in range(100,400)],[i for i in range(200,500)]]).T,columns=['a','b','c'])
s = setup(data, target='c')
# custom_loss = make_scorer(custom_loss)
add_metric('custom_loss', 'Custom Loss', custom_loss)
best = compare_models()
predict_model(best)

3、深度学习框架

（1）torch

python 复制代码

import torch
import torch.nn as nn
import numpy as np
import os

os.environ["KMP_DUPLICATE_LIB_OK"] = "TRUE"


def get_x_y():
    np.random.seed(0)
    x = np.random.randint(0, 50, 300)
    y_values = 2 * x + 21
    x = np.array(x, dtype=np.float32)
    y = np.array(y_values, dtype=np.float32)
    x = x.reshape(-1, 1)
    y = y.reshape(-1, 1)
    return x, y


class LinearRegressionModel(nn.Module):
    def __init__(self, input_dim, output_dim):
        super(LinearRegressionModel, self).__init__()
        self.linear = nn.Linear(input_dim, output_dim)  # 输入的个数，输出的个数

    def forward(self, x):
        out = self.linear(x)
        return out


class CustomLoss(nn.Module):
    def __init__(self):
        super(CustomLoss, self).__init__()
        self.mse_loss = nn.MSELoss()

    def forward(self, x, y):
        mse_loss = torch.mean(torch.pow((x - y), 2))
        return mse_loss


if __name__ == '__main__':
    input_dim = 1
    output_dim = 1
    x_train, y_train = get_x_y()

    model = LinearRegressionModel(input_dim, output_dim)
    epochs = 1000  # 迭代次数
    optimizer = torch.optim.SGD(model.parameters(), lr=0.001)
    # model_loss = nn.MSELoss() # 使用MSE作为loss
    model_loss = CustomLoss()  # 自定义loss
    # 开始训练模型
    for epoch in range(epochs):
        epoch += 1
        # 注意转行成tensor
        inputs = torch.from_numpy(x_train)
        labels = torch.from_numpy(y_train)
        # 梯度要清零每一次迭代
        optimizer.zero_grad()
        # 前向传播
        outputs: torch.Tensor = model(inputs)
        # 计算损失
        loss = model_loss(outputs, labels)
        # 返向传播
        loss.backward()
        # 更新权重参数
        optimizer.step()
        if epoch % 50 == 0:
            print('epoch {}, loss {}'.format(epoch, loss.item()))

参看：pytorch自定义loss损失函数_python_脚本之家