# Text2SQL Task T5-Base + Fine-tuning on Spider + Table Augumentation

This is our T5 model fine-tuned on Spider using a schema serialization, which includes a table description for injecting domain knowledge into T5
## Running the model

Inspired by the work done by [Picard](https://github.com/ElementAI/picard/) by adding a table description to the question and serialized schema:

```python
[question] | [db_id] | [table] : [column] ( [content] , [content] ) , [column] ( ... ) , [...] | [table] : ... | ... description * [table] : <meaning of table>; [table] : <meaning of table> ; ....  
```