RuntimeError: Expected object of type torch.DoubleTensor but found type torch.FloatTensor for argument #2 'weight'

Question

My input tensor is torch.DoubleTensor type. But I got the RuntimeError below:

RuntimeError: Expected object of type torch.DoubleTensor but found type torch.FloatTensor for argument #2 'weight'

I didn't specify the type of the weight explicitly(i.e. I did not init my weight by myself. The weight is created by pytorch). What will influence the type of weight in the forward process?

Thanks a lot!!

After I transfer the input type to FloatTensor by .float(). The code can process correctly. But I still don't know what if I want doubletensor type input...... — Eric Kani
– Eric Kani, Commented Mar 21, 2018 at 13:19

layog · Accepted Answer · 2018-03-23 14:52:43Z

39

The default type for weights and biases are torch.FloatTensor. So, you'll need to cast either your model to torch.DoubleTensor or cast your inputs to torch.FloatTensor. For casting your inputs you can do

X = X.float()

or cast your complete model to DoubleTensor as

model = model.double()

You can also set the default type for all tensors using

pytorch.set_default_tensor_type('torch.DoubleTensor')

It is better to convert your inputs to float rather than converting your model to double, because mathematical computations on double datatype is considerably slower on GPU.

edited Mar 23, 2018 at 14:52

answered Mar 22, 2018 at 15:31

layog

4,8111 gold badge31 silver badges32 bronze badges

Sign up to request clarification or add additional context in comments.

6 Comments

Eric Kani Over a year ago

Thanks for your answer very much!! BTW, I meet another problem about loading model. The IDE raise UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 918: ordinal not in range(128) error when I use model = torch.load("file.pth") (ps: the .pth file comes from somewhere, and it was not trained by myself. The pth file trained by myself can be loaded using torch.load) .Thanks again!

layog Over a year ago

That is probably because the model was created (and saved) using Python3, whose default encoding is utf-8, but you are using Python2. Add # -*- coding: utf-8 -*- at top of your python file. Also it is not a good practice to save and load the complete model. This can break in a number of ways. For proper serialization of models see the official post

Eric Kani Over a year ago

Thank you very much. But I am using python3... And I have added -*- coding: utf-8 -*- at top of my pyhton file. It doesn't work...

layog Over a year ago

Then you can try to load the model in python2. For more details you can refer to this issue on GitHub

tverrbjelke Over a year ago

for me with torch0.4.1 it is not pytorch.set(...) but torch.set(...)

|

Shital Shah · Accepted Answer · 2018-12-08 07:21:24Z

3

I was also receiving exact same error. The root cause turned out to be this statement in my data loading code:

t = t.astype(np.float)

Here np.float translates to 64-bit float which maps to DoubleTensor. So changing this to,

t = t.astype(np.float32)

solved the issue.

answered Dec 8, 2018 at 7:21

Shital Shah

69.9k21 gold badges258 silver badges202 bronze badges

Collectives™ on Stack Overflow

RuntimeError: Expected object of type torch.DoubleTensor but found type torch.FloatTensor for argument #2 'weight'

2 Answers 2

6 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

6 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related