What is the cause of this error?

[vrajjshah]

python scripts/reader/interactive.py 09/20/2018 12:46:08 PM: [ CUDA enabled (GPU -1) ] 09/20/2018 12:46:08 PM: [ Initializing model... ] 09/20/2018 12:46:08 PM: [ Loading model /home/vraj/DrQA/data/reader/single.mdl ] 09/20/2018 12:46:11 PM: [ Initializing tokenizer... ] Traceback (most recent call last): File "scripts/reader/interactive.py", line 55, in predictor.cuda() File "/home/vraj/DrQA/drqa/reader/predictor.py", line 142, in cuda self.model.cuda() File "/home/vraj/DrQA/drqa/reader/model.py", line 469, in cuda self.network = self.network.cuda() File "/home/vraj/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 216, in cuda return self._apply(lambda t: t.cuda(device)) File "/home/vraj/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 146, in _apply module._apply(fn) File "/home/vraj/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 146, in _apply module._apply(fn) File "/home/vraj/anaconda3/lib/python3.6/site-packages/torch/nn/modules/module.py", line 146, in _apply module._apply(fn) File "/home/vraj/anaconda3/lib/python3.6/site-packages/torch/nn/modules/rnn.py", line 123, in _apply self.flatten_parameters() File "/home/vraj/anaconda3/lib/python3.6/site-packages/torch/nn/modules/rnn.py", line 111, in flatten_parameters params = rnn.get_parameters(fn, handle, fn.weight_buf) File "/home/vraj/anaconda3/lib/python3.6/site-packages/torch/backends/cudnn/rnn.py", line 165, in get_parameters assert filter_dim_a.prod() == filter_dim_a[0] AssertionError

[ajfisch]

Are you using CUDNN 7.1 with PyTorch 0.3.1? https://github.com/pytorch/pytorch/issues/5667

[vrajjshah]

Yes I got the error solved , thank you. For people having the same error, please install PyTorch 0.3.1 with CUDNN 7.0. I got the project working with CUDA V9.1.85