Add evaluation results on the amazon config and test split of squadshifts
#8
by autoevaluator HF Staff - opened
README.md
CHANGED
|
@@ -27,6 +27,23 @@ model-index:
|
|
| 27 |
type: total
|
| 28 |
value: 11869
|
| 29 |
verified: true
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 30 |
---
|
| 31 |
|
| 32 |
# roberta-base for QA
|
|
|
|
| 27 |
type: total
|
| 28 |
value: 11869
|
| 29 |
verified: true
|
| 30 |
+
- task:
|
| 31 |
+
type: question-answering
|
| 32 |
+
name: Question Answering
|
| 33 |
+
dataset:
|
| 34 |
+
name: squadshifts
|
| 35 |
+
type: squadshifts
|
| 36 |
+
config: amazon
|
| 37 |
+
split: test
|
| 38 |
+
metrics:
|
| 39 |
+
- name: Exact Match
|
| 40 |
+
type: exact_match
|
| 41 |
+
value: 70.14
|
| 42 |
+
verified: true
|
| 43 |
+
- name: F1
|
| 44 |
+
type: f1
|
| 45 |
+
value: 83.3389
|
| 46 |
+
verified: true
|
| 47 |
---
|
| 48 |
|
| 49 |
# roberta-base for QA
|