jbrry Irish-BERT issues

jbrry / Irish-BERT

Repository to store helper scripts for creating an Irish BERT model.

Other

9 stars 0 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Include Irish subset of No Language Left Behind

#129 jowagner opened 4 months ago
0
Include Irish subset of Glot500

#128 jowagner opened 4 months ago
0
Include Irish subset of CulturaX

#127 jowagner opened 4 months ago
0
Include Irish subset of Leipzig Corpora Collection

#126 jowagner opened 4 months ago
0
Release of the Cloze Test set?

#125 KhanhTungTran closed 7 months ago
7
Repair fadas in NCI

#124 jowagner opened 1 year ago
0
Update HF model cards to refer to LREC paper

#123 jowagner closed 1 year ago
1
Add the classical modern Irish Corpas Filíocht shiollach na Gaeilge

#122 jowagner opened 1 year ago
0
Add the EduGA Corpus of Educational Materials

#121 jowagner opened 1 year ago
0
Add the Gaois Corpus of Contemporary Irish

#120 jowagner opened 1 year ago
0
Add Irish subset of Indigenous Blogs

#119 jowagner opened 1 year ago
0
Add the Irish Crúbadán Web Corpus

#118 jowagner opened 1 year ago
0
Repair ligatures in NCI

#117 jowagner opened 1 year ago
0
Use ELRC and OPUS corpora directly

#116 jowagner opened 1 year ago
0
Add Irish subset of Indigenous Tweets

#115 jowagner opened 1 year ago
0
Investigate gaHealth parallel corpus

#114 jowagner opened 1 year ago
0
add rclone --drive-shared-with-me flag for shared folder

#113 jowagner closed 2 years ago
1
rclone is unable to find Theme A folder on Google Driver

#112 jbrry closed 2 years ago
1
Reference for NCI paper Kilgarriff et al.

#111 jowagner closed 2 years ago
1
Include Scannell's corpus

#110 jowagner opened 2 years ago
0
report statistical power of test sets

#109 jowagner opened 2 years ago
0
tag, branch and/or release code for reproducibility

#108 jowagner opened 2 years ago
3
Effect of corpus sampling on continued pre-training

#107 jowagner opened 2 years ago
0
Increase number of parsers from 5 to 9

#106 jowagner opened 2 years ago
0
corpus statistics after de-duplication

#105 jowagner opened 2 years ago
0
Paper: report random seeds of from scratch models

#104 jowagner closed 2 years ago
1
Paper: include token counts / corpus stats

#103 jowagner closed 2 years ago
1
Paper: include other Irish BERT models in related work

#102 jowagner closed 2 years ago
1
Investigate the cross-lingual transferability of monolingual BERT representations to Irish

#101 jbrry opened 3 years ago
2
Multilingual model with Irish and Scottish Gaelic

#100 jowagner opened 3 years ago
0
Include transcription from the Nationale Folklore Collection

#99 jowagner opened 3 years ago
1
Mismatch between implementation and description of punctuation filter

#98 jowagner closed 2 years ago
2
Release corpora where possible

#97 jowagner opened 3 years ago
0
Upgrade to newest OSCAR release

#96 jowagner opened 3 years ago
0
Improve language filter

#95 jowagner opened 3 years ago
0
Role of top layers in fine-tuning BERT

#94 jowagner opened 3 years ago
0
Attach parser to lower layer of BERT

#93 jowagner opened 3 years ago
0
Add POS tagging as evaluation task

#92 jowagner opened 3 years ago
0
Effect of MT ouput in training corpora

#91 jowagner opened 3 years ago
0
Check prevalence of MT output in web-crawled corpora

#90 jowagner opened 3 years ago
0
Investigate mC4 dataset

#89 jbrry opened 3 years ago
0
Train a generative model such as GPT-2

#88 jbrry opened 3 years ago
0
Add more open, alternative license

#87 jowagner opened 3 years ago
0
Meta: Summary of future work ideas / feature requests

#86 jowagner opened 3 years ago
0
Effect of random initalisation of BERT from scratch models

#85 jowagner opened 3 years ago
0
Add an evaluation task more focused on semantics

#84 jowagner opened 3 years ago
0
Citation section in readme

#83 jowagner opened 3 years ago
0
Which effect of `document-filter` is responsible for improved LAS?

#82 jowagner opened 3 years ago
0
Why does electra trained for 24h not perform well?

#81 jowagner opened 3 years ago
0
Investigate effect of ## glue on prefixes

#80 jowagner opened 3 years ago
0