mxnet实现线性回归（linear regression）

拼搏现实的明天。 2022-05-12 05:14 198阅读 0赞

采用mxnet实现线性回归算法

'coding = utf-8'
    import mxnet.ndarray as nd
    from mxnet import autograd
    import random
    
    #数据的生成1000*2
    num_inputs = 2
    num_examples = 1000
    
    ture_w = [2,-3.4]
    ture_b = 4.2
    
    x = nd.random_normal(shape=(num_examples,num_inputs))
    y = ture_w[0] * x[:,0] + ture_w[1] * x[:,1] + ture_b
    y += .01 * nd.random_normal(shape=y.shape)
    print(x[0:10],y[0:10])
    
    #数据读取
    batch_size = 10
    def data_iter():
        #产生一个随机索引
        index = list(range(num_examples))
        random.shuffle(index)
        for i in range(0,num_examples,batch_size):
            j = nd.array(index[i:min(i + batch_size,num_examples)])
            yield nd.take(x,j),nd.take(y,j)
    
    for data ,label in data_iter():
        print(data,label)
        break
    
    #初始化参数
    w = nd.random_normal(shape=(num_inputs,1))
    b = nd.zeros((1,))
    params = [w,b]
    for param in params:
        param.attach_grad()
    
    #定义模型
    def net(x):
        return nd.dot(x,w) + b
    
    #损失函数
    def square_loss(yhat,y):
        #把y变为yhat避免自动广播
        return (yhat - y.reshape(yhat.shape)) ** 2
    #优化
    def SGD(params,lr):
        for param in params:
            param[:] = param - lr * param.grad
    
    #训练
    epochs = 5
    learning_rate = 0.001
    for e in range(epochs):
        total_loss = 0
        for data ,label in data_iter():
            with autograd.record():
                output = net(data)
                loss = square_loss(output,label)
            loss.backward()
            SGD(params,learning_rate)
            total_loss += nd.sum(loss).asscalar()
        print('Epoch %d,average loss:%f'%(e,total_loss/num_examples))
    
    
    print(ture_b,b)#验证学习的效果
    print(ture_w,w)

运行结果：

[[ 1.1630785   0.4838046 ]
     [ 0.29956347  0.15302546]
     [-1.1688148   1.558071  ]
     [-0.5459446  -2.3556297 ]
     [ 0.54144025  2.6785064 ]
     [ 1.2546344  -0.54877406]
     [-0.68106437 -0.1353156 ]
     [ 0.37723133  0.41016456]
     [ 0.5712682  -2.7579627 ]
     [ 1.07628    -0.6141325 ]]
    <NDArray 10x2 @cpu(0)> 
    [ 4.879625   4.2968144 -3.4331114 11.099875  -3.8235688  8.576558
      3.3012044  3.5644817 14.710667   8.445549 ]
    <NDArray 10 @cpu(0)>
    
    [[ 0.2508148  -0.30159083]
     [ 0.1703758   1.4159782 ]
     [ 0.60099405 -0.04193413]
     [ 0.5398333  -0.08873118]
     [ 0.5390026   0.15226784]
     [-0.92770946  0.6714997 ]
     [ 0.29670075  1.3111951 ]
     [-0.56099415 -0.9978546 ]
     [-0.64676666  1.3366979 ]
     [ 1.5275036  -1.3969318 ]]
    <NDArray 10x2 @cpu(0)> 
    [ 5.7448215  -0.26755202  5.538153    5.5854945   4.7548504   0.04230662
      0.34051183  6.4591236  -1.6538469  11.992856  ]
    <NDArray 10 @cpu(0)>
    Epoch 0,average loss:5.551557
    Epoch 1,average loss:0.098503
    Epoch 2,average loss:0.001866
    Epoch 3,average loss:0.000130
    Epoch 4,average loss:0.000097
    4.2 
    [4.2002177]
    <NDArray 1 @cpu(0)>
    [2, -3.4] 
    [[ 1.9998528]
     [-3.400118 ]]
    <NDArray 2x1 @cpu(0)>
    
    Process finished with exit code 0