crossroderick
/

aramt5

Text Generation

Classical Syriac

text2text-generation

transliteration

Eval Results (legacy)

Model card Files Files and versions

aramt5 / src /data

174 kB

Ctrl+K

Ctrl+K

1 contributor

History: 10 commits

crossroderick's picture

v3.2 hotfix with some corrections and new files

ace5912 about 1 month ago

augment_atomic_tokens.py

6.1 kB
Data augmentation and balancing updates for a re-run of v3 about 2 months ago
balance_corpus.py

10.5 kB
Data augmentation and balancing updates for a re-run of v3 about 2 months ago
build_corpus_vocabulary.py

5.88 kB
v3.1 updadte about 2 months ago
correction_dataset.jsonl

43.1 kB
xet

v3.2 hotfix with some corrections and new files about 1 month ago
fetch_sedra_vocalised.py

12.3 kB
Data augmentation and balancing updates for a re-run of v3 about 2 months ago
fetch_syriac_corpus.py

3.28 kB
Initial commit about 2 months ago
generate_clean_corpus.sh

952 Bytes
Initial commit about 2 months ago
generate_syr_lat_pairs.py

37.8 kB
v3.1 updadte about 2 months ago
get_data.sh

1.09 kB
Initial commit about 2 months ago
manual_vocabulary.jsonl

34.4 kB
xet

Manual vocabulary updates for upcoming v3.2 update about 2 months ago
sedra_lookup.py

18.9 kB
Manual vocabulary updates for upcoming v3.2 update about 2 months ago