Datasets:
Add Colab tutorial link
Browse files
README.md
CHANGED
|
@@ -14,6 +14,9 @@ configs:
|
|
| 14 |
---
|
| 15 |
# Structured Extraction
|
| 16 |
|
|
|
|
|
|
|
|
|
|
| 17 |
Documents paired with a JSON Schema and the JSON object extracted from them. The documents (emails, receipts, bookings, forms, letters, listings, ...) cover 35 document types in 16 languages. All people, companies, addresses and identifiers are fictional.
|
| 18 |
|
| 19 |
## Splits
|
|
|
|
| 14 |
---
|
| 15 |
# Structured Extraction
|
| 16 |
|
| 17 |
+
> [!Note]
|
| 18 |
+
> 📓 **Tutorial**: This is the dataset used in [SFT with TRL](https://colab.research.google.com/github/Liquid4All/cookbook/blob/main/finetuning/notebooks/lfm3_sft_with_trl.ipynb)
|
| 19 |
+
|
| 20 |
Documents paired with a JSON Schema and the JSON object extracted from them. The documents (emails, receipts, bookings, forms, letters, listings, ...) cover 35 document types in 16 languages. All people, companies, addresses and identifiers are fictional.
|
| 21 |
|
| 22 |
## Splits
|