深度學習&PyTorch筆記 (2) logistic 迴歸
阿新 • • 發佈:2018-12-13
logistic 適用於分類問題。 同之前引入所需要的庫
import torch
import torch.nn as nn
from torch.autograd import Variable
import numpy as np
import matplotlib.pyplot as plt
建立模型,使用Sigmoid函式
class LogisticRegression(nn.Module): def __init__(self): super(LogisticRegression, self).__init__() self.lr = nn.Linear(2, 1) self.sm = nn.Sigmoid() def forward(self, x): x = self.lr(x) x = self.sm(x) return x
例項化,選擇損失函式和優化函式。 由於選擇的是Sigmoid函式,所以損失函式選擇BCE,公式為
????=−(?∗???(?̂)+(1−?)∗???(1−?̂))loss=−(y∗log(y)+(1−y)∗log(1−y)) 優化方式還是選擇梯度下降法,找到損失函式的梯度,沿著梯度的方向進行優化,以便找到損失的極小值,來確定引數。
logistic_model = LogisticRegression() criterion = nn.BCELoss() optimizer = torch.optim.SGD(logistic_model.parameters(), lr=1e-3, omentum=0.9)
開始訓練
for epoch in range(50000): x = Variable(x_data) y = Variable(y_data) # forward out = logistic_model(x) loss = criterion(out, y) print_loss = loss.item() mask = out.ge(0.5).float() correct = (mask == y).sum() acc = correct.item() / x.size(0) # backward optimizer.zero_grad() loss.backward() optimizer.step()
最後的結果如下:
w0, w1 = logistic_model.lr.weight[0]
w0 = w0.item()
w1 = w1.item()
b0 = logistic_model.lr.bias.item()
plot_x = np.arange(30, 100, 0.1)
plot_y = (-w0 * plot_x - b0) / w1