I installed PyTorch on Windows following the Easy Install steps on https://github.com/peterjc123/pytorch-scripts
It works fine but when I am trying to use LSTM, it fails with NameError: name ‘fn’ is not defined.
I checked issues section of PyTorch https://github.com/pytorch/pytorch/issues/5211 and I understand that the bug is fixed in the master branch.
I have no clue how to build PyTorch from master branch on Windows for GPU version. I am new to this.
I am using Anaconda environment.
Any suggestions will be helpful. I am stuck!
Thanks in advance.
- OS: Windows 10
- PyTorch version: 0.3.1
- How you installed PyTorch (conda, pip, source): conda install -c peterjc123 pytorch cuda90
- Python version: 3.5
- CUDA/cuDNN version: 9
- GPU models and configuration: NVIDIA GTX 1060