Datasets:
Tasks:
Question Answering
Modalities:
Text
Formats:
json
Languages:
Japanese
Size:
10K - 100K
License:
Update README.md
Browse files
README.md
CHANGED
|
@@ -10,6 +10,8 @@ Each record has the following format :
|
|
| 10 |
|
| 11 |
in which Context is the context to input to an LLM for RAG evaluation.
|
| 12 |
|
|
|
|
|
|
|
| 13 |
|
| 14 |
|
| 15 |
|
|
@@ -30,6 +32,8 @@ LLMに入れるときに、下記のようなプロンプトで入れてくだ
|
|
| 30 |
段落: {Context}
|
| 31 |
```
|
| 32 |
|
|
|
|
|
|
|
| 33 |
|
| 34 |
---
|
| 35 |
license: cc-by-sa-4.0
|
|
|
|
| 10 |
|
| 11 |
in which Context is the context to input to an LLM for RAG evaluation.
|
| 12 |
|
| 13 |
+
Evaluation script :
|
| 14 |
+
https://github.com/ducalt/jcrrageval
|
| 15 |
|
| 16 |
|
| 17 |
|
|
|
|
| 32 |
段落: {Context}
|
| 33 |
```
|
| 34 |
|
| 35 |
+
自動評価スクリプト:
|
| 36 |
+
https://github.com/ducalt/jcrrageval
|
| 37 |
|
| 38 |
---
|
| 39 |
license: cc-by-sa-4.0
|