【项目实战典型案例】09.数据类型不一致导致equals判断为false

### 使用 PyTorch 进行深度学习的企业级项目实战案例 #### 1. 图像分类器构建 图像分类是计算机视觉中的基础任务之一。通过使用卷积神经网络 (CNN)，可以有效地识别和分类不同类型的图片。在企业环境中，这可用于产品自动分类、质量控制等领域。 ```python import torch from torchvision import datasets, transforms, models transform = transforms.Compose([transforms.Resize(255), transforms.CenterCrop(224), transforms.ToTensor()]) dataset = datasets.ImageFolder('path_to_images', transform=transform) dataloader = torch.utils.data.DataLoader(dataset, batch_size=64, shuffle=True) model = models.resnet50(pretrained=True) for param in model.parameters(): param.requires_grad = False classifier = nn.Sequential(OrderedDict([ ('fc1', nn.Linear(2048, 512)), ('relu', nn.ReLU()), ('dropout', nn.Dropout(p=0.2)), ('output', nn.Linear(512, num_classes))])) model.fc = classifier criterion = nn.CrossEntropyLoss() optimizer = optim.Adam(model.fc.parameters(), lr=0.003) device = torch.device("cuda" if torch.cuda.is_available() else "cpu") model.to(device); epochs = 5 steps = 0 running_loss = 0 print_every = 5 for epoch in range(epochs): for inputs, labels in dataloader: steps += 1 inputs, labels = inputs.to(device), labels.to(device) optimizer.zero_grad() logps = model.forward(inputs) loss = criterion(logps, labels) loss.backward() optimizer.step() running_loss += loss.item() if steps % print_every == 0: test_loss = 0 accuracy = 0 with torch.no_grad(): model.eval() for inputs, labels in validloader: inputs, labels = inputs.to(device), labels.to(device) logps = model.forward(inputs) batch_loss = criterion(logps, labels) test_loss += batch_loss.item() ps = torch.exp(logps) top_p, top_class = ps.topk(1, dim=1) equals = top_class == labels.view(*top_class.shape) accuracy += torch.mean(equals.type(torch.FloatTensor)).item() train_losses.append(running_loss/print_every) validation_losses.append(test_loss/len(validloader)) accuracies.append(accuracy/len(validloader)) print(f"Epoch {epoch+1}/{epochs}.. " f"Train loss: {running_loss/print_every:.3f}.. " f"Validation loss: {test_loss/len(validloader):.3f}.. " f"Accuracy: {(accuracy/len(validloader))*100}%") running_loss = 0 model.train() ``` 这段代码展示了如何利用预训练模型 ResNet50 来创建一个简单的图像分类器，并对其进行微调以适应特定的数据集[^1]。 #### 2. 自然语言处理应用——情感分析 自然语言处理(NLP) 是另一个广泛应用领域，在客户服务支持系统中尤为突出。下面是一个基于 LSTM 的文本情绪检测例子： ```python class SentimentRNN(nn.Module): def __init__(self, vocab_size, output_size, embedding_dim, hidden_dim, n_layers, drop_prob=0.5): super(SentimentRNN, self).__init__() self.output_size = output_size self.n_layers = n_layers self.hidden_dim = hidden_dim self.embedding = nn.Embedding(vocab_size, embedding_dim) self.lstm = nn.LSTM(embedding_dim, hidden_dim, n_layers, dropout=drop_prob, batch_first=True) self.dropout = nn.Dropout(drop_prob) self.fc = nn.Linear(hidden_dim, output_size) self.sigmoid = nn.Sigmoid() def forward(self, x, hidden): batch_size = x.size(0) embeds = self.embedding(x.long()) lstm_out, hidden = self.lstm(embeds, hidden) lstm_out = lstm_out.contiguous().view(-1, self.hidden_dim) out = self.dropout(lstm_out) out = self.fc(out) sig_out = self.sigmoid(out) sig_out = sig_out.view(batch_size, -1) sig_out = sig_out[:, -1] return sig_out, hidden def init_hidden(self, batch_size): weight = next(self.parameters()).data hidden = (weight.new(self.n_layers, batch_size, self.hidden_dim).zero_(), weight.new(self.n_layers, batch_size, self.hidden_dim).zero_()) return hidden ``` 此部分实现了循环神经网络(RNN)结构来预测电影评论的情感倾向（正面或负面），这对于监控在线平台上的公众舆论非常有用[^2]。 #### 3. 时间序列预测—股票价格走势建模 对于金融行业来说，时间序列分析至关重要。这里介绍了一个采用 GRU 单元的时间序列回归问题解决方案，用于估计未来几天内的股价变动趋势。 ```python def create_sequences(data, seq_length): xs = [] ys = [] for i in range(len(data)-seq_length-1): x = data[i:(i+seq_length)] y = data[i+seq_length] xs.append(x) ys.append(y) return np.array(xs), np.array(ys) train_x, train_y = create_sequences(train_data.values.reshape(-1,1), SEQ_LENGTH) valid_x, valid_y = create_sequences(val_data.values.reshape(-1,1), SEQ_LENGTH) gru_model = Sequential() gru_model.add(GRU(units=50, return_sequences=True, input_shape=(SEQ_LENGTH, 1))) gru_model.add(Dropout(0.2)) gru_model.add(GRU(units=50)) gru_model.add(Dropout(0.2)) gru_model.add(Dense(units=1)) gru_model.compile(optimizer='adam', loss='mean_squared_error') history = gru_model.fit(train_x, train_y, epochs=EPOCHS, batch_size=BATCH_SIZE, verbose=1, validation_split=0.1) ``` 上述片段说明了怎样运用门控循环单元(Gated Recurrent Unit,GRU)来进行短期股市行情预报，有助于投资者做出更明智的投资决策[^3].