Mihai-Dan MAŞALA (25095) commited on
Commit
7bd01ab
1 Parent(s): 364ef40

Update README

Browse files
Files changed (1) hide show
  1. README.md +6 -8
README.md CHANGED
@@ -69,33 +69,31 @@ Model | Dev | Test
69
  -------|---------|----------
70
  multilingual-BERT | 68.96 | 69.57
71
  XLM-R-base | 71.26 | 71.71
72
- [BERT-base-ro](https://huggingface.co/dumitrescustefan/bert-base-romanian-uncased-v1) | 70.49 | 71.02
73
  RoBERT-small | 66.32 | 66.37
74
  RoBERT-base | 70.89 | 71.61
75
  RoBERT-large | 72.48 | 72.11
76
 
77
  ### Moldavian vs. Romanian Dialect and Cross-dialect Topic identification
78
 
79
- We report results on [VarDial 2019](https://sites.google.com/view/vardial2019/campaign) Moldavian vs. Romanian Cross-dialect Topic identification Challenge, as Macro-averaged F1 score (in %)
80
 
81
  Model | Dialect Classification | MD to RO | RO to MD
82
- -------|---------|----------
83
  2-CNN + SVM | 93.40 | 65.09 | 75.21
84
  Char+Word SVM | 96.20 | 69.08 | 81.93
85
  BiGRU | 93.30 | 70.10 | 80.30
86
 
87
  multilingual-BERT | 95.34 | 68.76 | 78.24
88
  XLM-R-base | 96.28 | 69.93 | 8228
89
- [BERT-base-ro](https://huggingface.co/dumitrescustefan/bert-base-romanian-uncased-v1) | 96.20 | 69.93 | 78.79
90
  RoBERT-small | 95.67 | 69.01 | 80.40
91
  RoBERT-base | 97.39 | 68.30 | 81.09
92
  RoBERT-large | 97.78 | 69.91 | 83.65
93
 
94
  ### Diacritics Restoration
95
 
96
- Challenge can be found [here](https://diacritics-challenge.speed.pub.ro/).
97
-
98
- We report results on the official test set, as accuracies in %.
99
 
100
  Model | word level | char level
101
  -------|---------|----------
@@ -103,7 +101,7 @@ BiLSTM | 99.42 | -
103
  CharCNN | 98.40 | 99.65
104
  CharCNN + multilingual-BERT | 99.72 | 99.94
105
  CharCNN + XLM-R-base | 99.76 | 99.95
106
- CharCNN + [BERT-base-ro](https://huggingface.co/dumitrescustefan/bert-base-romanian-uncased-v1) | 99.79 | 99.95
107
  CharCNN + RoBERT-small | 99.73 | 99.94
108
  CharCNN + RoBERT-base | 99.78 | 99.95
109
  CharCNN + RoBERT-large | 99.76 | 99.95
69
  -------|---------|----------
70
  multilingual-BERT | 68.96 | 69.57
71
  XLM-R-base | 71.26 | 71.71
72
+ BERT-base-ro | 70.49 | 71.02
73
  RoBERT-small | 66.32 | 66.37
74
  RoBERT-base | 70.89 | 71.61
75
  RoBERT-large | 72.48 | 72.11
76
 
77
  ### Moldavian vs. Romanian Dialect and Cross-dialect Topic identification
78
 
79
+ We report results on [VarDial 2019](https://sites.google.com/view/vardial2019/campaign) Moldavian vs. Romanian Cross-dialect Topic identification Challenge, as Macro-averaged F1 score (in %).
80
 
81
  Model | Dialect Classification | MD to RO | RO to MD
82
+ -------|---------|----------|----------|
83
  2-CNN + SVM | 93.40 | 65.09 | 75.21
84
  Char+Word SVM | 96.20 | 69.08 | 81.93
85
  BiGRU | 93.30 | 70.10 | 80.30
86
 
87
  multilingual-BERT | 95.34 | 68.76 | 78.24
88
  XLM-R-base | 96.28 | 69.93 | 8228
89
+ BERT-base-ro | 96.20 | 69.93 | 78.79
90
  RoBERT-small | 95.67 | 69.01 | 80.40
91
  RoBERT-base | 97.39 | 68.30 | 81.09
92
  RoBERT-large | 97.78 | 69.91 | 83.65
93
 
94
  ### Diacritics Restoration
95
 
96
+ Challenge can be found [here](https://diacritics-challenge.speed.pub.ro/). We report results on the official test set, as accuracies in %.
 
 
97
 
98
  Model | word level | char level
99
  -------|---------|----------
101
  CharCNN | 98.40 | 99.65
102
  CharCNN + multilingual-BERT | 99.72 | 99.94
103
  CharCNN + XLM-R-base | 99.76 | 99.95
104
+ CharCNN + BERT-base-ro | 99.79 | 99.95
105
  CharCNN + RoBERT-small | 99.73 | 99.94
106
  CharCNN + RoBERT-base | 99.78 | 99.95
107
  CharCNN + RoBERT-large | 99.76 | 99.95