Add evaluation results on the 3.0.0 config of cnn_dailymail
#5
by
autoevaluator
HF staff
- opened
README.md
CHANGED
|
@@ -2,6 +2,108 @@
|
|
| 2 |
language: en
|
| 3 |
tags:
|
| 4 |
- summarization
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
---
|
| 6 |
|
| 7 |
### Pegasus Models
|
|
|
|
| 2 |
language: en
|
| 3 |
tags:
|
| 4 |
- summarization
|
| 5 |
+
model-index:
|
| 6 |
+
- name: google/pegasus-xsum
|
| 7 |
+
results:
|
| 8 |
+
- task:
|
| 9 |
+
type: summarization
|
| 10 |
+
name: Summarization
|
| 11 |
+
dataset:
|
| 12 |
+
name: samsum
|
| 13 |
+
type: samsum
|
| 14 |
+
config: samsum
|
| 15 |
+
split: train
|
| 16 |
+
metrics:
|
| 17 |
+
- name: ROUGE-1
|
| 18 |
+
type: rouge
|
| 19 |
+
value: 21.8096
|
| 20 |
+
verified: true
|
| 21 |
+
- name: ROUGE-2
|
| 22 |
+
type: rouge
|
| 23 |
+
value: 4.2525
|
| 24 |
+
verified: true
|
| 25 |
+
- name: ROUGE-L
|
| 26 |
+
type: rouge
|
| 27 |
+
value: 17.4469
|
| 28 |
+
verified: true
|
| 29 |
+
- name: ROUGE-LSUM
|
| 30 |
+
type: rouge
|
| 31 |
+
value: 18.8907
|
| 32 |
+
verified: true
|
| 33 |
+
- name: loss
|
| 34 |
+
type: loss
|
| 35 |
+
value: 3.0317161083221436
|
| 36 |
+
verified: true
|
| 37 |
+
- name: gen_len
|
| 38 |
+
type: gen_len
|
| 39 |
+
value: 20.3122
|
| 40 |
+
verified: true
|
| 41 |
+
- task:
|
| 42 |
+
type: summarization
|
| 43 |
+
name: Summarization
|
| 44 |
+
dataset:
|
| 45 |
+
name: xsum
|
| 46 |
+
type: xsum
|
| 47 |
+
config: default
|
| 48 |
+
split: test
|
| 49 |
+
metrics:
|
| 50 |
+
- name: ROUGE-1
|
| 51 |
+
type: rouge
|
| 52 |
+
value: 46.8623
|
| 53 |
+
verified: true
|
| 54 |
+
- name: ROUGE-2
|
| 55 |
+
type: rouge
|
| 56 |
+
value: 24.4533
|
| 57 |
+
verified: true
|
| 58 |
+
- name: ROUGE-L
|
| 59 |
+
type: rouge
|
| 60 |
+
value: 39.0548
|
| 61 |
+
verified: true
|
| 62 |
+
- name: ROUGE-LSUM
|
| 63 |
+
type: rouge
|
| 64 |
+
value: 39.0994
|
| 65 |
+
verified: true
|
| 66 |
+
- name: loss
|
| 67 |
+
type: loss
|
| 68 |
+
value: 1.5717021226882935
|
| 69 |
+
verified: true
|
| 70 |
+
- name: gen_len
|
| 71 |
+
type: gen_len
|
| 72 |
+
value: 22.8821
|
| 73 |
+
verified: true
|
| 74 |
+
- task:
|
| 75 |
+
type: summarization
|
| 76 |
+
name: Summarization
|
| 77 |
+
dataset:
|
| 78 |
+
name: cnn_dailymail
|
| 79 |
+
type: cnn_dailymail
|
| 80 |
+
config: 3.0.0
|
| 81 |
+
split: test
|
| 82 |
+
metrics:
|
| 83 |
+
- name: ROUGE-1
|
| 84 |
+
type: rouge
|
| 85 |
+
value: 22.2062
|
| 86 |
+
verified: true
|
| 87 |
+
- name: ROUGE-2
|
| 88 |
+
type: rouge
|
| 89 |
+
value: 7.6701
|
| 90 |
+
verified: true
|
| 91 |
+
- name: ROUGE-L
|
| 92 |
+
type: rouge
|
| 93 |
+
value: 15.4046
|
| 94 |
+
verified: true
|
| 95 |
+
- name: ROUGE-LSUM
|
| 96 |
+
type: rouge
|
| 97 |
+
value: 19.2182
|
| 98 |
+
verified: true
|
| 99 |
+
- name: loss
|
| 100 |
+
type: loss
|
| 101 |
+
value: 2.681241273880005
|
| 102 |
+
verified: true
|
| 103 |
+
- name: gen_len
|
| 104 |
+
type: gen_len
|
| 105 |
+
value: 25.0234
|
| 106 |
+
verified: true
|
| 107 |
---
|
| 108 |
|
| 109 |
### Pegasus Models
|