mehrad
a77ecfdbb4
fix lgtm
2020-04-13 22:00:30 -07:00
mehrad
eb5bac57cc
fix assertion
2020-04-13 17:50:53 -07:00
mehrad
370828cbcf
more updates
2020-04-13 03:26:09 -07:00
mehrad
83ba114c1b
allow paired example training
2020-04-13 00:23:06 -07:00
Mehrad Moradshahi
88c3e83de8
Merge pull request #11 from stanford-oval/wip/multilanguage
...
Updates from Wip/multilanguage
2020-04-11 14:25:13 -07:00
mehrad
891bee8632
merge master branch
2020-04-09 22:10:35 -07:00
mehrad
97ea3c2fd2
fixes
2020-04-09 22:06:49 -07:00
Mehrad Moradshahi
a91947a755
Merge pull request #14 from stanford-oval/wip/features
...
Wip/features
2020-04-06 21:53:12 -07:00
mehrad
64df5dd6ac
remove default value
2020-04-06 18:15:11 -07:00
mehrad
31190c3f18
update tests
2020-04-06 16:03:05 -07:00
mehrad
93882915e9
allow choosing dev set for tuning
2020-04-06 15:54:44 -07:00
mehrad
11b84192e4
warn when datasets are not parallel
2020-04-06 15:36:02 -07:00
Giovanni Campagna
68f46d7568
v0.2.0
2020-04-03 10:28:04 -07:00
Giovanni Campagna
484d4904cf
Update Pipfile.lock
2020-04-03 10:26:44 -07:00
Giovanni Campagna
ffaf123b5c
v0.2.0b2
2020-04-02 18:49:14 -07:00
mehrad
73deed2a67
Merge branch 'master' into wip/multilanguage
2020-04-01 23:39:18 -07:00
mehrad
2f5330d9c1
fix bug
2020-04-01 23:20:49 -07:00
Mehrad Moradshahi
4a32a22ab6
fix switched prediction splits ( #13 )
2020-04-01 17:59:25 -07:00
Giovanni Campagna
39dd4aa5ad
Fix server for multilingual_eval ( #12 )
2020-04-01 17:59:04 -07:00
mehrad
70fc430a42
add assertion to avoid nan in std
2020-04-01 16:47:32 -07:00
mehrad
74401aa03f
use encoder_loss for training only
2020-04-01 02:10:38 -07:00
mehrad
f97f027b67
fix lgtm alert
2020-03-30 16:35:44 -07:00
mehrad
f3bf392231
adding tests + fixing bugs
2020-03-30 14:43:03 -07:00
mehrad
f57b43b14c
adding encoder loss
2020-03-30 13:54:47 -07:00
mehrad
703edc4fdc
Option to do sentence batching
...
Batches can have multiple minibatches; each containing same sentences but in different languages
2020-03-30 12:50:02 -07:00
mehrad
c97b219233
Merge branch 'master' into wip/multilanguage
2020-03-29 14:10:22 -07:00
mehrad
181a8208f2
Merge branch 'master' into wip/multilanguage
2020-03-29 14:09:35 -07:00
Giovanni Campagna
8d8d816777
Try fixing the docker build on dockerhub
2020-03-27 15:55:33 -07:00
Giovanni Campagna
5ca27fb5f7
v0.2.0b1
2020-03-27 13:08:42 -07:00
Giovanni Campagna
dbeb5a4fdd
Update Pipfile.lock
2020-03-27 13:08:34 -07:00
Mehrad Moradshahi
50fa36f354
Merge pull request #10 from stanford-oval/wip/multilanguage
...
Wip/multilanguage
2020-03-26 12:40:04 -07:00
mehrad
60adad8173
adding tests + bug fixes
2020-03-25 23:51:04 -07:00
mehrad
3bf3719edd
addressing pr comments
2020-03-25 17:43:04 -07:00
mehrad
10b759f568
move general data util files to scripts dir
2020-03-25 11:36:20 -07:00
mehrad
d870b634db
remove obsolete local data
2020-03-25 11:31:38 -07:00
mehrad
a2d2e740de
adding multilingual task
2020-03-25 02:33:19 -07:00
mehrad
87536fe2cb
allow caching multiple transformer embeddings
2020-03-24 19:52:42 -07:00
Giovanni Campagna
6810668b94
Post release version bump
2020-03-24 19:24:35 -07:00
Giovanni Campagna
02d9003539
v0.2.0a2
2020-03-24 18:55:59 -07:00
Giovanni Campagna
36b1197c9a
numericalizer/transformer: remove bogus assertions ( #9 )
...
These assertions do not mean much, because those tokens are guaranteed
to be in the decoder vocabulary regardless of the assertion, and
they won't necessarily have the same ID in the decoder and the true
vocabulary. Also the mask_id assertion fails for XLMR, because
mask_id is 250004.
2020-03-24 18:54:44 -07:00
mehrad
0acf64bdc3
address mask_id issue for XLM-R
2020-03-24 16:05:01 -07:00
Giovanni Campagna
1e2dbce017
Fix loading embeddings with untied embeddings ( #8 )
...
If embeddings for context & questions are untied with the "@" suffix,
we must not pass the suffix to the transformer library.
2020-03-23 00:56:41 -07:00
Giovanni Campagna
5a72ac7ff6
Post-release version bump
2020-03-21 19:31:38 -07:00
Giovanni Campagna
65338cb05d
v0.2.0a1
2020-03-21 19:08:19 -07:00
Giovanni Campagna
bb6018ba01
Add Pipfile script
...
So you can run "pipenv run genienlp" instead of "pipenv run python3 -m genienlp"
2020-03-19 10:37:16 -07:00
Giovanni Campagna
f39cfac2a6
Merge pull request #6 from stanford-oval/wip/contextual
...
First batch of changes from dialogue work
2020-03-19 10:04:42 -07:00
Giovanni Campagna
2106ef1cb0
Merge pull request #7 from stanford-oval/wip/export
...
Add "genienlp export" command
2020-03-17 20:49:44 -07:00
Giovanni Campagna
123ea6802b
Add "genienlp export" command
...
The command copies over the model files that are needed for inference,
without intermediate checkpoints.
2020-03-17 15:42:26 -07:00
Giovanni Campagna
69e6707773
Fix tests
2020-03-16 13:17:23 -07:00
Giovanni Campagna
f971c31dde
Merge remote-tracking branch 'origin/master' into wip/contextual
2020-03-16 12:42:21 -07:00