RuntimeError: "argmax_cuda" not implemented for 'Bool'

autograd
1

I am getting this error when I run the following code:

score_flattened = score_.view(B * NP * SQ, B2 * NS * SQ)
target_flattened = target_.view(B * NP * SQ, B2 * NS * SQ)
target_flattened = target_flattened.argmax(dim=1)

This error only happens on Google Colab but not on my own laptop (which has windows 10 on it)

2

Which version are you using locally and on Google Colab?
Could you update the Colab version, if it’s not the latest release?

3

Updated the colab version but I still get the same issue. Nevermind, I fixed the issue but just recasting my tensor as a float (before it was uint8)

4

I hit the same issue for the booleans, but not for bytes/uint8.
This is my environment:

Collecting environment information...
PyTorch version: 1.3.1
Is debug build: No
CUDA used to build PyTorch: 10.0.130

OS: Ubuntu 18.04.3 LTS
GCC version: (Ubuntu 7.4.0-1ubuntu1~18.04.1) 7.4.0
CMake version: Could not collect

Python version: 3.7
Is CUDA available: Yes
CUDA runtime version: Could not collect
GPU models and configuration: GPU 0: GeForce GTX TITAN X
Nvidia driver version: 418.40.04
cuDNN version: Could not collect

Versions of relevant libraries:
[pip] numpy==1.17.3
[pip] torch==1.3.1
[pip] torchvision==0.4.2
[conda] blas                      1.0                         mkl  
[conda] mkl                       2019.4                      243  
[conda] mkl-service               2.3.0            py37he904b0f_0  
[conda] mkl_fft                   1.0.15           py37ha843d7b_0  
[conda] mkl_random                1.1.0            py37hd6b4f25_0  
[conda] pytorch                   1.3.1           py3.7_cuda10.0.130_cudnn7.6.3_0    pytorch
[conda] torchvision               0.4.2                py37_cu100    pytorch

I don’t have the issue with the pytorch 1.2/cuda 10.0 combo.

5

Recast your tensor to float:

my_tensor = my_tensor.double()

1 Like
6

I found this useful: RuntimeError: “argmax_cuda” not implemented for ‘Bool’