Update README.md
Browse files
README.md
CHANGED
|
@@ -1 +1,26 @@
|
|
| 1 |
-
#
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# t5_wikisql_SQL2en
|
| 2 |
+
|
| 3 |
+
This is the checkpoint for [t5_wikisql_SQL2en](https://huggingface.co/dbernsohn/t5_wikisql_SQL2en) after being fune-tunedon the wikisql dataset on t5-small.
|
| 4 |
+
|
| 5 |
+
The model can be loaded like so:
|
| 6 |
+
|
| 7 |
+
```python
|
| 8 |
+
from transformers import AutoModelWithLMHead, AutoTokenizer
|
| 9 |
+
|
| 10 |
+
tokenizer = AutoTokenizer.from_pretrained("t5-small")
|
| 11 |
+
model = AutoModelWithLMHead.from_pretrained("/content/model")
|
| 12 |
+
```
|
| 13 |
+
|
| 14 |
+
You can then use this model to translate SQL queries into plain english.
|
| 15 |
+
|
| 16 |
+
```python
|
| 17 |
+
input_text = f"translate English to SQL: {query} </s>"
|
| 18 |
+
features = tokenizer([input_text], return_tensors='pt')
|
| 19 |
+
|
| 20 |
+
output = model.generate(input_ids=features['input_ids'],
|
| 21 |
+
attention_mask=features['attention_mask'])
|
| 22 |
+
|
| 23 |
+
tokenizer.decode(output[0])
|
| 24 |
+
```
|
| 25 |
+
|
| 26 |
+
The whole training process and hyperparameters are in my []GitHub repo] (https://github.com/DorBernsohn)
|