深度学习_10_softmax

深度学习_10_softmax_实战

由于网上代码的画图功能是基于jupyter记事本，而我用的是pycham,这导致画图代码不兼容pycharm,所以删去部分代码，以便能更好的在pycharm上运行

完整代码：

import torch
from d2l import torch as d2l

"创建训练集&创建检测集合"
batch_size = 256
train_iter, test_iter = d2l.load_data_fashion_mnist(batch_size)

"创建模型w, b"
num_inputs = 784
num_outputs = 10

W = torch.normal(0, 0.01, size=(num_inputs, num_outputs), requires_grad=True)
b = torch.zeros(num_outputs, requires_grad=True)

"softmax"
def softmax(X):
    X_exp = torch.exp(X)
    partition = X_exp.sum(1, keepdim=True)
    return X_exp / partition  # 这里应用了广播机制

"输出，即传入图片输出"
def net(X):
    return softmax(torch.matmul(X.reshape((-1, W.shape[0])), W) + b)

"交叉熵损失"
def cross_entropy(y_hat, y):
    return - torch.log(y_hat[range(len(y_hat)), y])

"显示预测与估计相对应下标数量"
def accuracy(y_hat, y):  #@save
    """计算预测正确的数量"""
    if len(y_hat.shape) > 1 and y_hat.shape[1] > 1: # 确定长宽高都大于1
        y_hat = y_hat.argmax(axis=1) # 取出每行中最大值
    cmp = y_hat.type(y.dtype) == y
    return float(cmp.type(y.dtype).sum()) # 返回对应下标数量


"利用优化后的模型计算精度"
def evaluate_accuracy(net, data_iter):  #@save

    if isinstance(net, torch.nn.Module):
        net.eval()  # 将模型设置为评估模式
    metric = Accumulator(2)  # 正确预测数、预测总数
    with torch.no_grad():
        for X, y in data_iter:
            metric.add(accuracy(net(X), y), y.numel()) # 下标相同数量 / 总下标
    return metric[0] / metric[1]


"加法器"
class Accumulator:  #@save

    def __init__(self, n):
        self.data = [0.0] * n

    def add(self, *args):
        self.data = [a + float(b) for a, b in zip(self.data, args)]

    def reset(self):
        self.data = [0.0] * len(self.data)

    def __getitem__(self, idx):
        return self.data[idx]

"训练更新模型&返回训练损失与精度函数"
def train_epoch_ch3(net, train_iter, loss, updater):  #@save
    """训练模型一个迭代周期（定义见第3章）"""
    # 将模型设置为训练模式
    if isinstance(net, torch.nn.Module):
        net.train()
    # 训练损失总和、训练准确度总和、样本数
    metric = Accumulator(3)
    for X, y in train_iter:
        # 计算梯度并更新参数
        y_hat = net(X)
        l = loss(y_hat, y)
        if isinstance(updater, torch.optim.Optimizer):
            # 使用PyTorch内置的优化器和损失函数
            updater.zero_grad()
            l.mean().backward()
            updater.step()
        else:
            # 使用定制的优化器和损失函数
            l.sum().backward()
            updater(X.shape[0])
        metric.add(float(l.sum()), accuracy(y_hat, y), y.numel())
    # 返回训练损失和训练精度
    return metric[0] / metric[2], metric[1] / metric[2]

lr = 0.1

"更新模型"
def updater(batch_size):
    return d2l.sgd([W, b], lr, batch_size)

if __name__ == '__main__':
    num_epochs = 10
    cnt = 1
    for i in range(num_epochs):
        X, Y = train_epoch_ch3(net, train_iter, cross_entropy, updater)
        print("训练次数: " + str(cnt))
        cnt += 1
        print("训练损失: {:.4f}".format(X))
        print("训练精度: {:.4f}".format(Y))
        print(".................................")
#        print(W)
#        print(b)