Add evaluation results on the default config of xsum
#5
by
autoevaluator
HF staff
- opened
README.md
CHANGED
@@ -1,11 +1,78 @@
|
|
1 |
---
|
2 |
tags:
|
3 |
- summarization
|
4 |
-
|
5 |
language:
|
6 |
- en
|
7 |
-
|
8 |
license: mit
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
9 |
---
|
10 |
### Bart model finetuned on xsum
|
11 |
|
|
|
1 |
---
|
2 |
tags:
|
3 |
- summarization
|
|
|
4 |
language:
|
5 |
- en
|
|
|
6 |
license: mit
|
7 |
+
model-index:
|
8 |
+
- name: facebook/bart-large-xsum
|
9 |
+
results:
|
10 |
+
- task:
|
11 |
+
type: summarization
|
12 |
+
name: Summarization
|
13 |
+
dataset:
|
14 |
+
name: cnn_dailymail
|
15 |
+
type: cnn_dailymail
|
16 |
+
config: 3.0.0
|
17 |
+
split: test
|
18 |
+
metrics:
|
19 |
+
- name: ROUGE-1
|
20 |
+
type: rouge
|
21 |
+
value: 25.2697
|
22 |
+
verified: true
|
23 |
+
- name: ROUGE-2
|
24 |
+
type: rouge
|
25 |
+
value: 7.6638
|
26 |
+
verified: true
|
27 |
+
- name: ROUGE-L
|
28 |
+
type: rouge
|
29 |
+
value: 17.1808
|
30 |
+
verified: true
|
31 |
+
- name: ROUGE-LSUM
|
32 |
+
type: rouge
|
33 |
+
value: 21.7933
|
34 |
+
verified: true
|
35 |
+
- name: loss
|
36 |
+
type: loss
|
37 |
+
value: 3.5042972564697266
|
38 |
+
verified: true
|
39 |
+
- name: gen_len
|
40 |
+
type: gen_len
|
41 |
+
value: 27.4462
|
42 |
+
verified: true
|
43 |
+
- task:
|
44 |
+
type: summarization
|
45 |
+
name: Summarization
|
46 |
+
dataset:
|
47 |
+
name: xsum
|
48 |
+
type: xsum
|
49 |
+
config: default
|
50 |
+
split: test
|
51 |
+
metrics:
|
52 |
+
- name: ROUGE-1
|
53 |
+
type: rouge
|
54 |
+
value: 45.4525
|
55 |
+
verified: true
|
56 |
+
- name: ROUGE-2
|
57 |
+
type: rouge
|
58 |
+
value: 22.3455
|
59 |
+
verified: true
|
60 |
+
- name: ROUGE-L
|
61 |
+
type: rouge
|
62 |
+
value: 37.2302
|
63 |
+
verified: true
|
64 |
+
- name: ROUGE-LSUM
|
65 |
+
type: rouge
|
66 |
+
value: 37.2323
|
67 |
+
verified: true
|
68 |
+
- name: loss
|
69 |
+
type: loss
|
70 |
+
value: 2.3128726482391357
|
71 |
+
verified: true
|
72 |
+
- name: gen_len
|
73 |
+
type: gen_len
|
74 |
+
value: 25.5435
|
75 |
+
verified: true
|
76 |
---
|
77 |
### Bart model finetuned on xsum
|
78 |
|