Pythorch自定义丢失功能

Question

如何实现自定义丢失功能？使用以下代码会导致错误：

import torch
import torch.nn as nn
import torchvision
import torchvision.transforms as transforms
import numpy as np
import matplotlib.pyplot as plt
import torch.utils.data as data_utils
import torch.nn as nn
import torch.nn.functional as F

num_epochs = 20

x1 = np.array([0,0])
x2 = np.array([0,1])
x3 = np.array([1,0])
x4 = np.array([1,1])

num_epochs = 200

class cus2(torch.nn.Module):

    def __init__(self):
        super(cus2,self).__init__()

    def forward(self, outputs, labels):
        # reshape labels to give a flat vector of length batch_size*seq_len
        labels = labels.view(-1)  

        # mask out 'PAD' tokens
        mask = (labels >= 0).float()

        # the number of tokens is the sum of elements in mask
        num_tokens = int(torch.sum(mask).data[0])

        # pick the values corresponding to labels and multiply by mask
        outputs = outputs[range(outputs.shape[0]), labels]*mask

        # cross entropy loss for all non 'PAD' tokens
        return -torch.sum(outputs)/num_tokens


x = torch.tensor([x1,x2,x3,x4]).float()

y = torch.tensor([0,1,1,0]).long()

train = data_utils.Tensordataset(x,y)
train_loader = data_utils.DataLoader(train , batch_size=2 , shuffle=True)

device = 'cpu'

input_size = 2
hidden_size = 100 
num_classes = 2

learning_rate = .0001

class NeuralNet(nn.Module) : 
    def __init__(self, input_size, hidden_size, num_classes) : 
        super(NeuralNet, self).__init__()
        self.fc1 = nn.Linear(input_size , hidden_size)
        self.relu = nn.ReLU()
        self.fc2 = nn.Linear(hidden_size , num_classes)

    def forward(self, x) : 
        out = self.fc1(x)
        out = self.relu(out)
        out = self.fc2(out)
        return out

for i in range(0 , 1) :

        model = NeuralNet(input_size, hidden_size, num_classes).to(device)

        criterion = nn.CrossEntropyLoss()
#         criterion = Regress_Loss()
#         criterion = cus2()
        optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

        total_step = len(train_loader)
        for epoch in range(num_epochs) : 
            for i,(images , labels) in enumerate(train_loader) : 
                images = images.reshape(-1 , 2).to(device)
                labels = labels.to(device)

                outputs = model(images)
                loss = criterion(outputs , labels)

                optimizer.zero_grad()
                loss.backward()
                optimizer.step()
#                 print(loss)

        outputs = model(x)

        print(outputs.data.max(1)[1])

对培训数据进行完美预测：

tensor([0, 1, 1, 0])

使用来自https://cs230-stanford.github.io/pytorch-nlp.html#writing-a-custom-loss-function的自定义丢失函数：

在上面的代码中实现为cus2

取消注释代码# criterion = cus2()以使用此loss函数返回：

tensor([0, 0, 0, 0])

同时返回警告：

UserWarning: invalid index of a 0-dim tensor. This will be an error in PyTorch 0.5. Use tensor.item() to convert a 0-dim tensor to a Python number

我没有正确实现自定义丢失功能？

Answer 1

您的loss函数在编程上是正确的，但以下情况除外：

    # the number of tokens is the sum of elements in mask
    num_tokens = int(torch.sum(mask).data[0])

当您执行torch.sum时，它将返回一个0维张量，并因此发出无法索引的警告。若要修复此问题，请按建议执行int(torch.sum(mask).item())，否则int(torch.sum(mask))也将起作用。

现在，你是想用定制损耗来模拟CE损耗吗？如果是，那么您就缺少log_softmax

要修复此问题，请在语句4之前添加outputs = torch.nn.functional.log_softmax(outputs, dim=1)。注意，在附加教程的情况下，log_softmax已经在forward调用中完成。你也可以这么做。

另外，我注意到学习速度慢，甚至与CE丢失，结果不一致。把学习率提高到1e-3对我来说很好，以防客户和CE丢失。