Datasets:
YoussefJoo
commited on
Commit
•
8fc91b9
1
Parent(s):
208dc11
Update README.md
Browse files
README.md
CHANGED
@@ -11,4 +11,37 @@ size_categories:
|
|
11 |
task_categories:
|
12 |
- text2text-generation
|
13 |
pretty_name: OGText2SQL
|
14 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
task_categories:
|
12 |
- text2text-generation
|
13 |
pretty_name: OGText2SQL
|
14 |
+
---
|
15 |
+
|
16 |
+
# Dataset Summary
|
17 |
+
**OGText2SQL** dataset was utilized in training the **OGSQL** model, this dataset comprises over 350,000 rows of text-to-SQL pairs. Through a series of data refining steps, including schema expansion, SQL refinement, and instruction generation using existing Language Models (LLMs), the dataset was meticulously processed to ensure quality and relevance.
|
18 |
+
|
19 |
+
# How to use it
|
20 |
+
|
21 |
+
> Python
|
22 |
+
```python
|
23 |
+
from datasets import load_dataset
|
24 |
+
|
25 |
+
dataset = load_dataset("OneGate/OGText2SQL")
|
26 |
+
```
|
27 |
+
|
28 |
+
|
29 |
+
> API
|
30 |
+
```shell
|
31 |
+
curl -X GET \
|
32 |
+
"https://datasets-server.huggingface.co/rows?dataset=OneGate%2FOGText2SQL&config=default&split=train&offset=0&length=100"
|
33 |
+
```
|
34 |
+
|
35 |
+
# Dataset Structure
|
36 |
+
## Data Instances
|
37 |
+
Each data instance in this dataset represents a text-to-SQL entry where the instruction has been formatted with the table schema and question. The output is the SQL in SQlite dialect.
|
38 |
+
|
39 |
+
## Data Fields
|
40 |
+
|
41 |
+
- `context`: the instruction to generate SQL.
|
42 |
+
- `question`: the human input.
|
43 |
+
- `SQL`: the generated SQL query.
|
44 |
+
- `source`: the source dataset of the sample.
|
45 |
+
|
46 |
+
## Languages
|
47 |
+
The language of the data is mainly English, with other languages.
|