Add evaluation results on the HadiPourmousa--TextSummarization config and train split of HadiPourmousa/TextSummarization
#1
by
autoevaluator
HF staff
- opened
README.md
CHANGED
@@ -21,6 +21,39 @@ model-index:
|
|
21 |
- name: Rouge1
|
22 |
type: rouge
|
23 |
value: 24.7231
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
---
|
25 |
|
26 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
|
|
21 |
- name: Rouge1
|
22 |
type: rouge
|
23 |
value: 24.7231
|
24 |
+
- task:
|
25 |
+
type: summarization
|
26 |
+
name: Summarization
|
27 |
+
dataset:
|
28 |
+
name: HadiPourmousa/TextSummarization
|
29 |
+
type: HadiPourmousa/TextSummarization
|
30 |
+
config: HadiPourmousa--TextSummarization
|
31 |
+
split: train
|
32 |
+
metrics:
|
33 |
+
- name: ROUGE-1
|
34 |
+
type: rouge
|
35 |
+
value: 5.2877
|
36 |
+
verified: true
|
37 |
+
- name: ROUGE-2
|
38 |
+
type: rouge
|
39 |
+
value: 1.2836
|
40 |
+
verified: true
|
41 |
+
- name: ROUGE-L
|
42 |
+
type: rouge
|
43 |
+
value: 5.0791
|
44 |
+
verified: true
|
45 |
+
- name: ROUGE-LSUM
|
46 |
+
type: rouge
|
47 |
+
value: 5.086
|
48 |
+
verified: true
|
49 |
+
- name: loss
|
50 |
+
type: loss
|
51 |
+
value: 4.567324161529541
|
52 |
+
verified: true
|
53 |
+
- name: gen_len
|
54 |
+
type: gen_len
|
55 |
+
value: 19.0
|
56 |
+
verified: true
|
57 |
---
|
58 |
|
59 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|