issues
search
mead-ml
/
mead-baseline
Deep-Learning Model Exploration and Development for NLP
Apache License 2.0
243
stars
73
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Fix collection class references
#933
khatchad
opened
1 year ago
0
CVE-2007-4559 Patch
#932
TrellixVulnTeam
opened
2 years ago
0
WIP: Feature/hydra
#931
dpressel
opened
2 years ago
0
check for invalid input in classify file, discard
#930
dpressel
closed
2 years ago
0
Fix/onnx recent pytorch
#929
dpressel
closed
2 years ago
0
support distillation with soft targets
#928
dpressel
closed
2 years ago
0
Chore/add gpt2 completer
#927
dpressel
closed
2 years ago
0
gpt2 to npz conversion
#926
shar999
closed
2 years ago
0
fix issue in postLN encoder block (tf only)
#925
dpressel
closed
2 years ago
0
refactor transformer layer -> easier understanding
#924
dpressel
closed
2 years ago
0
Implemented the joint bert model
#923
shar999
closed
2 years ago
0
hotfix for label vec
#922
dpressel
closed
2 years ago
0
Feature/jointbert
#921
shar999
closed
2 years ago
1
add support for classification with dual encoder
#920
dpressel
closed
2 years ago
0
support JSON output
#919
dpressel
closed
2 years ago
0
in N-best should be dynamic on dim 1
#918
dpressel
closed
2 years ago
0
remove dyn axis on len
#917
dpressel
closed
2 years ago
0
hotfix for multifeature input with nbest
#916
dpressel
closed
2 years ago
0
WIP: Feature/mlm repl
#915
dpressel
opened
2 years ago
0
add option to not zip checkpoint
#914
dpressel
closed
2 years ago
0
minor mod to support loading all features
#913
dpressel
closed
2 years ago
0
use the setup.cfg to do the lifting
#912
dpressel
closed
2 years ago
0
Feature/add offsets to spm
#911
dpressel
closed
2 years ago
0
TF docs say we should shard upfront
#910
dpressel
closed
2 years ago
0
For known args, we should add them as lists once
#909
dpressel
closed
2 years ago
0
Fix/sp chars spm
#908
dpressel
closed
2 years ago
0
Feature/tf grad accum tlm
#907
dpressel
closed
2 years ago
0
Feature/ra updates
#906
dpressel
closed
2 years ago
0
Chore/improve spm int
#905
dpressel
closed
2 years ago
0
escape
#904
dpressel
closed
2 years ago
0
fix line delim
#903
dpressel
closed
2 years ago
0
Chore/docker ext
#902
dpressel
closed
2 years ago
0
Chore/gpt2 tag vecs
#901
dpressel
closed
2 years ago
0
add regex to docker image for GPT style vecs
#900
dpressel
closed
2 years ago
0
Chore/add cu111
#899
dpressel
closed
2 years ago
0
add df for cu111
#898
dpressel
closed
2 years ago
0
fix wordpiece secondary feats to use transform_fn
#897
dpressel
closed
2 years ago
0
fix secondary BPE featurizer
#896
dpressel
closed
2 years ago
0
fixes for stand-alone DE prog
#895
dpressel
closed
2 years ago
0
rm -e so mead-train works as expected
#894
dpressel
closed
2 years ago
0
Revert "try and fix entrypoint issue"
#893
dpressel
closed
2 years ago
0
try and fix entrypoint issue
#892
dpressel
closed
2 years ago
0
add options to classify text for labels and fmt
#891
dpressel
closed
2 years ago
0
fixing triton wrong output issue
#890
shar999
closed
2 years ago
1
Add ability to specify extra text2embedding cols
#889
dpressel
closed
2 years ago
0
Feature/cossim loss
#888
dpressel
closed
2 years ago
0
dont weight decay LN or biases
#887
dpressel
closed
2 years ago
0
Chore/sep paired training
#886
dpressel
closed
2 years ago
0
WIP: switch to ONNX compat torch-only tril
#885
dpressel
closed
2 years ago
0
test sentinels and only append the token if not
#884
dpressel
closed
2 years ago
0
Next