Update README.md
Browse files
README.md
CHANGED
@@ -5,4 +5,28 @@ datasets:
|
|
5 |
language:
|
6 |
- en
|
7 |
pipeline_tag: text-classification
|
8 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
language:
|
6 |
- en
|
7 |
pipeline_tag: text-classification
|
8 |
+
---
|
9 |
+
|
10 |
+
# kgourgou/bert-base-uncased-QA-classification
|
11 |
+
|
12 |
+
An experiment into classifying whether a pair of (question, answer) is valid.
|
13 |
+
|
14 |
+
Input must be formatted as
|
15 |
+
|
16 |
+
```
|
17 |
+
question: {your query}? answer: {your possible answer}
|
18 |
+
```
|
19 |
+
|
20 |
+
The output probabilities are for
|
21 |
+
|
22 |
+
1. class 0 = the answer string couldn't be an answer to the question and
|
23 |
+
2. class 1 = the answer string could be an answer to the question.
|
24 |
+
|
25 |
+
|
26 |
+
Examples:
|
27 |
+
|
28 |
+
- "question: What number comes after five? answer: four" -> this should be class 1 as the answer is a number (even if it's not the right number).
|
29 |
+
- "question: Which person is associated with Kanye West? answer: a tree" -> this should be class 0 as a tree is not a person.
|
30 |
+
|
31 |
+
|
32 |
+
|