Register
Login
Resources
Docs Blog Datasets Glossary Case Studies Tutorials & Webinars
Product
Data Engine LLMs Platform Enterprise
Pricing Explore
Connect to our Discord channel
Myle Ott 7e86e30cc5
Allow larger maxlen (fixes #100) (#101)
6 years ago
..
66314a60d8
ATen Fix
6 years ago
f9362e87bd
Output correct perplexity when training with --sentence-avg
6 years ago
c5378602d4
Share input/output embed
6 years ago
907ca927eb
Better support for torch.no_grad (since volatile is deprecated)
6 years ago
9a95121633
Adjust weight decay by the current learning rate to make it work correctly during annealing
6 years ago
cb0d7b2ad1
Fix flake8 warnings
6 years ago
1ff3efce63
Ratio should be predlen/reflen not reflen/predlen
6 years ago
334694363b
Better warning message for inputs that are too long
6 years ago
dcbf5e7533
Raise FileNotFoundError if dictionary files don't exist
6 years ago
42a0150c37
Replace unk with original string
6 years ago
d74f200aa8
Fixed 2 typos (#75)
6 years ago
e734b0fa58
Initial commit
6 years ago
e734b0fa58
Initial commit
6 years ago
ee36a6f3e3
Fixed Weight Decay Regularization in Adam
6 years ago
376c265f35
Add support for NCCL v2
6 years ago
c5378602d4
Share input/output embed
6 years ago
884e30464b
Update requirements.txt and fix flake8 (#62)
6 years ago
71d2d44c80
Prepare scripts for WMT14
6 years ago
66d9fcf5c8
Fix tests
6 years ago

Comments

Loading...