Comments (7)
Could you please share the error? Where does it occur?
from cpc.
OK, I think, you need to move hidden variable generated at line 101 of src/model.py to the device by using self.device.
For example:
hidden = th.zeros(2, bs, self.options["conv_dims"][-1])
hidden = hidden.to(self.device)
OR you can edit line 101 as following:
hidden = th.zeros((2, bs, self.options["conv_dims"][-1]), device=self.device)
So, either way should work. Please try, and see if it works. And if it does, please let me know.
from cpc.
Great! I will update the source code sometime soon.
Just a side note, in line 43 of ./utils/load_data.py
, test_dataset is commented out.
If you want to use the test data, download "dev-clean" from original source where you downloaded the training set, and uncomment line 43. In this case, you need to change line 45 from return train_dataset, train_dataset
to return train_dataset, test_dataset
I am closing this issue.
from cpc.
Hi, thanks for responding.
Traceback (most recent call last):
File "0_train.py", line 72, in <module>
main(config)
File "0_train.py", line 53, in main
train(config, data_loader, save_weights=True)
File "0_train.py", line 29, in train
encoder.fit(data_loader)
File "/home/mattias/PyFlow_CPC/src/model.py", line 103, in fit
encoder_samples, predictions, hidden = self.cpc(Xbatch, hidden)
File "/home/mattias/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
File "/home/mattias/PyFlow_CPC/utils/model_utils.py", line 60, in forward
output, hidden = self.gru(prior_sequence, hidden)
File "/home/mattias/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
File "/home/mattias/anaconda3/lib/python3.8/site-packages/torch/nn/modules/rnn.py", line 837, in forward
result = _VF.gru(input, hx, self._flat_weights, self.bias, self.num_layers,
RuntimeError: Input and hidden tensors are not at the same device, found input tensor at cuda:0 and hidden tensor at cpu
from cpc.
Getting further now but an error at:
Traceback (most recent call last):
File "0_train.py", line 72, in <module>
main(config)
File "0_train.py", line 53, in main
train(config, data_loader, save_weights=True)
File "0_train.py", line 29, in train
encoder.fit(data_loader)
File "/home/mattias/PyFlow_CPC/src/model.py", line 122, in fit
_ = self.validate(Xval) if epoch % self.options["nth_epoch"] == 0 else None
File "/home/mattias/PyFlow_CPC/src/model.py", line 189, in validate
encoder_samples, predictions, hidden = self.cpc(Xval, hidden)
File "/home/mattias/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
File "/home/mattias/PyFlow_CPC/utils/model_utils.py", line 60, in forward
output, hidden = self.gru(prior_sequence, hidden)
File "/home/mattias/anaconda3/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1051, in _call_impl
return forward_call(*input, **kwargs)
File "/home/mattias/anaconda3/lib/python3.8/site-packages/torch/nn/modules/rnn.py", line 837, in forward
result = _VF.gru(input, hx, self._flat_weights, self.bias, self.num_layers,
RuntimeError: Input and hidden tensors are not at the same device, found input tensor at cuda:0 and hidden tensor at cpu
from cpc.
You have to do the same thing at line 187.
# S=num_layers*num_directions, i.e. 2*1 since we are using 2 layers, and GRU is set as uni-directional
hidden = th.zeros(2, self.options["batch_size"], self.options["conv_dims"][-1])
from cpc.
Worked great now! Thanks for the input.
from cpc.
Related Issues (2)
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from cpc.