Find the error when running the train_hbmp.sh & train.py about hbmp HOT 9 CLOSED

Could be related to PyTorch version. The code is written using PyTorch 0.3.1. We have not tested with 0.4.

from hbmp.

Hi. Thanks for using our code. Are you running this on a machine with a GPU? Which version of PyTorch are you using. -Aarne

from hbmp.

Yes, I am running this on Google Colab and selecting GPU on Jupyter notebook.
PyTorch-0.4.1

from hbmp.

Now I have these versions:
Python (3.6.7)
Pytorch (0.3.1)
Numpy (1.14.3)
Torchtext (0.2.1)
SpaCy (2.0.11)
But there is an error:

Traceback (most recent call last):
File "train.py", line 393, in
main()
File "train.py", line 156, in main
map_location=lambda storage, location: storage.cuda(config.gpu))
File "/usr/local/lib/python3.6/dist-packages/torch/serialization.py", line 267, in load
return _load(f, map_location, pickle_module)
File "/usr/local/lib/python3.6/dist-packages/torch/serialization.py", line 420, in _load
result = unpickler.load()
AttributeError: Can't get attribute '_rebuild_tensor_v2' on <module 'torch._utils' from '/usr/local/lib/python3.6/dist-packages/torch/_utils.py'>

from hbmp.

Could it be that the code cannot locate the pretrained word embeddings. Have you run the download_data.sh script? That should download the word embeddings.

from hbmp.

Yes, I have.

from hbmp.

I tried to solve this issue:
https://github.com/Zeinab-Kouhkan/HBMP

from hbmp.

Hi, Unfortunately I don't have resources or time to verify and fix the issue. Let me know if you find a fix.
Aarne

from hbmp.

Closing this for now. Let me know if you still experience the issue.

from hbmp.