Add evaluation results on masakhaner dataset

#1
by autoevaluator HF staff - opened
Files changed (1) hide show
  1. README.md +35 -2
README.md CHANGED
@@ -1,5 +1,5 @@
1
  ---
2
- language:
3
  - swa
4
  tags:
5
  - NER
@@ -11,7 +11,40 @@ metrics:
11
  - recall
12
  license: apache-2.0
13
  widget:
14
- - text: "Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa, watu takriban 14 zaidi wamepata maambukizi ya Covid-19."
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
15
  ---
16
 
17
  # Model description
 
1
  ---
2
+ language:
3
  - swa
4
  tags:
5
  - NER
 
11
  - recall
12
  license: apache-2.0
13
  widget:
14
+ - text: Wizara ya afya ya Tanzania imeripoti Jumatatu kuwa, watu takriban 14 zaidi
15
+ wamepata maambukizi ya Covid-19.
16
+ model-index:
17
+ - name: arnolfokam/mbert-base-uncased-ner-swa
18
+ results:
19
+ - task:
20
+ type: token-classification
21
+ name: Token Classification
22
+ dataset:
23
+ name: masakhaner
24
+ type: masakhaner
25
+ config: swa
26
+ split: test
27
+ metrics:
28
+ - name: Accuracy
29
+ type: accuracy
30
+ value: 0.9478152826672189
31
+ verified: true
32
+ - name: Precision
33
+ type: precision
34
+ value: 0.956198880757641
35
+ verified: true
36
+ - name: Recall
37
+ type: recall
38
+ value: 0.9524405059672694
39
+ verified: true
40
+ - name: F1
41
+ type: f1
42
+ value: 0.9543159929827074
43
+ verified: true
44
+ - name: loss
45
+ type: loss
46
+ value: 0.38853195309638977
47
+ verified: true
48
  ---
49
 
50
  # Model description