Afshin Rahimi
Afshin Rahimi
Shouldn't instead of ` log_liklihoods.append(output[:, target]) ` We have `log_liklihoods.append(torch.gather(output, dim=1, index=target.unsqueeze(-1)))` ? Assume target size (batch size) is 64 and output is 64x4 (4 classes), output[:, target] gives me...
I'm using the sparse module, the code works for a small dataset and also works for a large dataset where deterministic = True (no dropout). But when I use dropout...
Thanks f0k (and nouiz), I'll try the suggestions.
Hi, That conversion was done for privacy and by the original publishers of the dataset, so we can't get the real handle from it. The datasets are quite old, so...
Solved my problem as well. Thanks.
seems you need to install scipy. you can install scipy by "pip install scipy".
The evaluation for the test set of the corresponding dataset is reported in: Rahimi, Afshin, Trevor Cohn, and Timothy Baldwin. "pigeo: A python geotagging tool." Proceedings of ACL-2016 System Demonstrations...
What Python version are you using? This was written for Python 2 not Python 3.
Hi, Memory requirement: small for GeoText (you can run it on any laptop) big for the other two (you'll need about 30GB RAM). Run Time: few minutes for GeoText, 3...
Hi, Sorry for the delay. the pickle file contains the parameters of the classifier (a GraphConv instance) after it is trained, and can be loaded by that instance using GraphConvinstance.load...