Add evaluation results on the squad_v2 config of squad_v2
#1
by
abhishek
HF staff
- opened
README.md
CHANGED
@@ -3,6 +3,66 @@ language: en
|
|
3 |
datasets:
|
4 |
- squad_v2
|
5 |
license: cc-by-4.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
---
|
7 |
|
8 |
# roberta-base for QA
|
|
|
3 |
datasets:
|
4 |
- squad_v2
|
5 |
license: cc-by-4.0
|
6 |
+
model-index:
|
7 |
+
- name: autoevaluate/roberta-base-squad2
|
8 |
+
results:
|
9 |
+
- task:
|
10 |
+
type: question-answering
|
11 |
+
name: Question Answering
|
12 |
+
dataset:
|
13 |
+
name: squad_v2
|
14 |
+
type: squad_v2
|
15 |
+
config: squad_v2
|
16 |
+
split: validation
|
17 |
+
metrics:
|
18 |
+
- name: Exact Match
|
19 |
+
type: exact_match
|
20 |
+
value: 79.9309
|
21 |
+
verified: true
|
22 |
+
- name: F1
|
23 |
+
type: f1
|
24 |
+
value: 82.9433
|
25 |
+
verified: true
|
26 |
+
- name: exact
|
27 |
+
type: exact
|
28 |
+
value: 79.9309
|
29 |
+
verified: true
|
30 |
+
- name: f1
|
31 |
+
type: f1
|
32 |
+
value: 82.9433
|
33 |
+
verified: true
|
34 |
+
- name: total
|
35 |
+
type: total
|
36 |
+
value: 11869
|
37 |
+
verified: true
|
38 |
+
- name: HasAns_exact
|
39 |
+
type: HasAns_exact
|
40 |
+
value: 79.9309
|
41 |
+
verified: true
|
42 |
+
- name: HasAns_f1
|
43 |
+
type: HasAns_f1
|
44 |
+
value: 82.9433
|
45 |
+
verified: true
|
46 |
+
- name: HasAns_total
|
47 |
+
type: HasAns_total
|
48 |
+
value: 11869
|
49 |
+
verified: true
|
50 |
+
- name: best_exact
|
51 |
+
type: best_exact
|
52 |
+
value: 79.9309
|
53 |
+
verified: true
|
54 |
+
- name: best_exact_thresh
|
55 |
+
type: best_exact_thresh
|
56 |
+
value: 0.0
|
57 |
+
verified: true
|
58 |
+
- name: best_f1
|
59 |
+
type: best_f1
|
60 |
+
value: 82.9433
|
61 |
+
verified: true
|
62 |
+
- name: best_f1_thresh
|
63 |
+
type: best_f1_thresh
|
64 |
+
value: 0.0
|
65 |
+
verified: true
|
66 |
---
|
67 |
|
68 |
# roberta-base for QA
|