Datasets:
Update README.md
Browse files
README.md
CHANGED
@@ -1,4 +1,16 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
dataset_info:
|
3 |
- config_name: mined-negative
|
4 |
features:
|
@@ -72,3 +84,62 @@ configs:
|
|
72 |
- split: train
|
73 |
path: triplet/train-*
|
74 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
multilinguality:
|
5 |
+
- monolingual
|
6 |
+
size_categories:
|
7 |
+
- 1M<n<10M
|
8 |
+
tags:
|
9 |
+
- sentence-transformers
|
10 |
+
task_categories:
|
11 |
+
- feature-extraction
|
12 |
+
- sentence-similarity
|
13 |
+
pretty_name: SQL Questions
|
14 |
dataset_info:
|
15 |
- config_name: mined-negative
|
16 |
features:
|
|
|
84 |
- split: train
|
85 |
path: triplet/train-*
|
86 |
---
|
87 |
+
|
88 |
+
# Dataset card for SQL Questions
|
89 |
+
|
90 |
+
This dataset is a reformatting of the [`sql_questions_triplets`](https://huggingface.co/datasets/sergeyvi4ev/sql_questions_triplets) dataset by [@sergeyvi4ev](https://huggingface.co/sergeyvi4ev), such that the dataset can be directly used to train Sentence Transformer models.
|
91 |
+
|
92 |
+
## Dataset Subsets
|
93 |
+
|
94 |
+
### `pair` subset
|
95 |
+
|
96 |
+
* Columns: "query", "positive"
|
97 |
+
* Column types: `str`, `str`
|
98 |
+
* Examples:
|
99 |
+
```python
|
100 |
+
{
|
101 |
+
'query': 'How many zip codes are under Barre, VT?',
|
102 |
+
'positive': '"Barre, VT" is the CBSA_name',
|
103 |
+
}
|
104 |
+
```
|
105 |
+
* Collection strategy: Reading the SQL Questions dataset and selecting all query-positive pairs.
|
106 |
+
* Deduplified: Yes
|
107 |
+
|
108 |
+
### `triplet` subset
|
109 |
+
|
110 |
+
* Columns: "query", "positive", "negative"
|
111 |
+
* Column types: `str`, `str`, `str`
|
112 |
+
* Examples:
|
113 |
+
```python
|
114 |
+
{
|
115 |
+
'query': 'How many zip codes are under Barre, VT?',
|
116 |
+
'positive': '"Barre, VT" is the CBSA_name',
|
117 |
+
'negative': "coordinates refers to latitude, longitude; latitude = '18.090875; longitude = '-66.867756'"
|
118 |
+
}
|
119 |
+
```
|
120 |
+
* Collection strategy: Reading the SQL Questions dataset and selecting all possible triplet pairs.
|
121 |
+
* Deduplified: No
|
122 |
+
|
123 |
+
### `mined-negative` subset
|
124 |
+
|
125 |
+
* Columns: "query", "positive", "negative_1", "negative_2", "negative_3", "negative_4", "negative_5", "negative_6", "negative_7", "negative_8", "negative_9", "negative_10"
|
126 |
+
* Column types: `str`, `str`, `str`, `str`, `str`, `str`, `str`, `str`, `str`, `str`, `str`, `str`
|
127 |
+
* Examples:
|
128 |
+
```python
|
129 |
+
{
|
130 |
+
"query": "How many zip codes are under Barre, VT?",
|
131 |
+
"positive": "\"Barre, VT\" is the CBSA_name",
|
132 |
+
"negative_1": "coordinates refers to latitude, longitude; latitude = '18.090875; longitude = '-66.867756'",
|
133 |
+
"negative_2": "name of county refers to county",
|
134 |
+
"negative_3": "median age over 40 refers to median_age > 40",
|
135 |
+
"negative_4": "\"PHILLIPS\" is the county; 'Montana' is the name of state",
|
136 |
+
"negative_5": "name of the CBSA officer refers to CBSA_name; position of the CBSA officer refers to CBSA_type;",
|
137 |
+
"negative_6": "population greater than 10000 in 2010 refers to population_2010 > 10000;",
|
138 |
+
"negative_7": "postal points refer to zip_code; under New York-Newark-Jersey City, NY-NJ-PA refers to CBSA_name = 'New York-Newark-Jersey City, NY-NJ-PA';",
|
139 |
+
"negative_8": "the largest water area refers to MAX(water_area);",
|
140 |
+
"negative_9": "\"Wisconsin\" is the state; largest land area refers to Max(land_area); full name refers to first_name, last_name; postal code refers to zip_code",
|
141 |
+
"negative_10": "\"Alabama\" and \"Illinois\" are both state; Ratio = Divide (Count(state = 'Alabama'), Count(state = 'Illinois'))"
|
142 |
+
}
|
143 |
+
```
|
144 |
+
* Collection strategy: Reading the SQL Questions dataset, filtering away the 15 samples that did not have 10 negative pairs, and formatting them in the described columns.
|
145 |
+
* Deduplified: No
|