RuntimeError: Expected tensor for argument #1 'indices' to have one of the following scalar types: Long, Int; but got torch.FloatTensor instead (while checking arguments for embedding)

这个错误是由于在使用embedding函数时，传递给indices参数的数据类型应该是Long或Int类型的tensor，但是你传递了一个FloatTensor类型的数据导致的。解决这个问题的方法是将传递给indices参数的数据类型转换为Long或Int类型的tensor，例如：

import torch

# 创建一个FloatTensor类型的数据
input_data = torch.randn(3, 4)

# 将数据类型转换为Long类型的tensor
indices = torch.LongTensor([[0, 1], [2, 0]])

# 使用embedding函数
embedding_layer = torch.nn.Embedding(10, 3)
output = embedding_layer(indices)

在上面的例子中，我们将传递给indices参数的数据类型从FloatTensor转换为LongTensor类型，这样就可以避免上述错误的出现。

相关问题

RuntimeError: Input type (torch.cuda.LongTensor) and weight type (torch.cuda.FloatTensor) should be the same

This error occurs when you are trying to perform an operation on a tensor and its corresponding weight, but their data types do not match. In this specific case, you are trying to perform an operation on a tensor of type `torch.cuda.LongTensor` and its corresponding weight of type `torch.cuda.FloatTensor`.

To resolve this error, you need to ensure that the data types of the tensor and its weight are the same. You can do this by converting one of them to match the other.

For example, if you want to convert the weight to type `torch.cuda.LongTensor`, you can use the `.long()` method to convert it:

weight = weight.long()

Alternatively, if you want to convert the tensor to type `torch.cuda.FloatTensor`, you can use the `.float()` method to convert it:

tensor = tensor.float()

Once you have converted one of the variables to match the other, you should be able to perform the operation without any issues.