时间戳

首页 > 解决方案 > 从经过训练的卷积网络进行预测

neural-network - 从经过训练的卷积网络进行预测

问题描述

这是我的网络，它创建训练数据，然后使用convolution单个激活来训练这些数据：convolutionrelu

train_dataset = []
mu, sigma = 0, 0.1 # mean and standard deviation
num_instances = 10

for i in range(num_instances) :
    image = []
    image_x = np.random.normal(mu, sigma, 1000).reshape((1 , 100, 10))
    train_dataset.append(image_x)

mu, sigma = 100, 0.80 # mean and standard deviation
for i in range(num_instances) :
    image = []
    image_x = np.random.normal(mu, sigma, 1000).reshape((1 , 100, 10))
    train_dataset.append(image_x)

labels_1 = [1 for i in range(num_instances)]
labels_0 = [0 for i in range(num_instances)]

labels = labels_1 + labels_0

print(labels)

x2 = torch.tensor(train_dataset).float()
y2 = torch.tensor(labels).long()

my_train2 = data_utils.TensorDataset(x2, y2)
train_loader2 = data_utils.DataLoader(my_train2, batch_size=batch_size_value, shuffle=False)


import torch 
import torch.nn as nn
import torchvision
import torchvision.transforms as transforms


# Device configuration
device = torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')
# device = 'cpu'

# Hyper parameters
num_epochs = 50
num_classes = 2
batch_size = 5
learning_rate = 0.001

# Convolutional neural network (two convolutional layers)
class ConvNet(nn.Module):
    def __init__(self, num_classes=1):
        super(ConvNet, self).__init__()
        self.layer1 = nn.Sequential(
            nn.Conv2d(1, 16, kernel_size=5, stride=1, padding=2),
            nn.BatchNorm2d(16),
            nn.ReLU(),
            nn.MaxPool2d(kernel_size=2, stride=2))
        self.layer2 = nn.Sequential(
            nn.Conv2d(16, 32, kernel_size=5, stride=1, padding=2),
            nn.BatchNorm2d(32),
            nn.ReLU(),
            nn.MaxPool2d(kernel_size=2, stride=2))
        self.fc = nn.Linear(32*25*2, num_classes)

    def forward(self, x):
        out = self.layer1(x)
        out = self.layer2(out)
        out = out.reshape(out.size(0), -1)
        out = self.fc(out)
        return out

model = ConvNet(num_classes).to(device)

# Loss and optimizer
criterion = nn.CrossEntropyLoss()
optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

# Train the model
total_step = len(train_loader2)
for epoch in range(num_epochs):
    for i, (images, labels) in enumerate(train_loader2):
        images = images.to(device)
        labels = labels.to(device)

        # Forward pass
        outputs = model(images)
        loss = criterion(outputs, labels)

        # Backward and optimize
        optimizer.zero_grad()
        loss.backward()
        optimizer.step()

        if (i % 10) == 0:
            print ('Epoch [{}/{}], Step [{}/{}], Loss: {:.4f}' 
                   .format(epoch+1, num_epochs, i+1, total_step, loss.item()))

为了做出一个单一的预测，我使用：

model(x2[10].unsqueeze_(0).cuda())

哪个输出：

tensor([[ 4.4880, -4.3128]], device='cuda:0')

这不应该返回预测的形状 (100,10) 的图像张量吗？

更新：为了执行预测，我使用：

torch.argmax(model(x2[2].unsqueeze_(0).cuda()), dim=1)

源代码：https ://discuss.pytorch.org/t/argmax-with-pytorch/1528/11

torch.argmax在这种情况下，返回使预测最大化的值的位置。

标签： neural-networkdeep-learningcomputer-visionconv-neural-networkpytorch

解决方案

正如Koustav所指出的，您的网络不是“完全卷积的”：尽管您有nn.Conv2d两层，但顶部仍然有一个“完全连接”（aka nn.Linear）层，它仅输出二维（num_classes）输出张量。

更具体地说，您的网络需要 1x100x10 输入（单通道，100 x 10 像素图像）。
在self.layer1你有一个 16x50x5 张量（来自卷积的 16 个通道，最大池化层减少了空间维度）之后。
在self.layer2你有一个 32x25x2 张量（来自卷积的 32 个通道，空间维度被另一个最大池化层减少）之后。
最后，您的全连接self.fc nn.Linear层采用整个维度的输入张量并从整个输入中32*25*2产生输出。num_classes

推荐阅读