Add world_language_model transformer and lstm #1005
Add examples submodule
2f234723
Add wlm transformer model
779383b3
Fix batchify code
1b987c42
Fix a few bugs
53ee00fc
Tune number of batches to run
0a4e6f34
Add per-batch latency
2a4b929c
Add compatible code
483be0bf
Add another fix
4ba34ce5
Fix v2 bisection ci
29d0f344
Add psutil
a85d13f0
Remove torchtext model
2cb20ef7
Remove Background_matting
2a091ac7
Remove LearningToPaint
d51008e2
Add metadata and install
5a0dff95
Remove torchtext from installer
35499324
Fix mps attr
49e07c70
Fix metadata
270b1f1c
Fix per-batch latency
9ba8ead2
Only run 5 batches
02e4b9fe
Run only 3 batches
cdbd3fe2
Skip installing benchmark
8164890c
Disable update repos, disable cuda build
50b9f526
Add back models
b061faf7
Add back attention
74d18ff8
xuzhao9
force pushed
from
fef120e8
to
74d18ff8
3 years ago
Fix v2 workflow
23b13832
Revert utils
36e1cf19
Add back csv
355653e9
Fix torchtext
6a3b2ddd
Reduce cpu test time
36895f94
Fix a bug
9ef40148
Add wlm_lstm model
53f91caf
Fix a bug
9b2d8c9c
Fix comments
7b7cff64
Fix a few bugs
bb1a9260
xuzhao9
changed the title Add world_language_model transformer Add world_language_model transformer and lstm 3 years ago
TODO: fix cuda memory leak
6f7fce0a
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub