Ouaill commited on
Commit
bdf5a6e
·
verified ·
1 Parent(s): ec48621

Upload doda_independent_results.csv with huggingface_hub

Browse files
Files changed (1) hide show
  1. doda_independent_results.csv +13 -0
doda_independent_results.csv ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ name,source,algorithm,architecture,vocab_size,n_texts,fertility,cpt,exact_match
2
+ concat_bpe_8000,ours,bpe,concatenated,8000,10000,2.2493365838654813,2.687371473272056,0.9997
3
+ concat_wordpiece_16000,ours,wordpiece,concatenated,16000,10000,2.0290824072414515,2.9837349169014926,0.9997
4
+ concat_bpe_32000,ours,bpe,concatenated,32000,10000,1.9088338778378482,3.1857039546146164,0.9997
5
+ CaMeLBERT-MSA,external_msa,WordPiece,shared,30000,10000,3.666749920577788,1.5885688345700721,0.9848
6
+ Asafaya-BERT,external_msa,WordPiece,shared,32000,10000,3.1872837153369504,1.8326136998271592,0.946
7
+ Aranizer-SP-86k,external_msa,SentencePiece,shared,86000,10000,3.0218238644117323,1.932848411795582,0.9871
8
+ B2BERT,external_msa,WordPiece,shared,30000,10000,3.666749920577788,1.5885688345700721,0.9848
9
+ DarijaBERT-ar,external_darija,WordPiece,shared,80000,10000,2.7214379379770555,2.17268830784525,0.9445
10
+ DarijaBERT-az,external_darija,WordPiece,shared,110000,10000,1.3138639463640935,4.552913171884771,0.9445
11
+ Moroccan-Darija-Tokenizer,external_darija,BPE,shared,30000,10000,3.328279298136504,1.7596048679100837,0.0003
12
+ Translit-Darija,external_darija,BPE,shared,30000,10000,1.869362422920217,3.206997841360009,0.0302
13
+ Qwen2.5-Darija,external_darija,SentencePiece,shared,151643,10000,2.6158528813017052,2.255188783435299,1.0