Instructions to use Sennodipoi/LayoutLMv2-FUNSD-ft with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Sennodipoi/LayoutLMv2-FUNSD-ft with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="Sennodipoi/LayoutLMv2-FUNSD-ft")# Load model directly from transformers import AutoProcessor, AutoModelForTokenClassification processor = AutoProcessor.from_pretrained("Sennodipoi/LayoutLMv2-FUNSD-ft") model = AutoModelForTokenClassification.from_pretrained("Sennodipoi/LayoutLMv2-FUNSD-ft", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Commit ·
a13079f
1
Parent(s): dbf0204
Update README.md
Browse files
README.md
CHANGED
|
@@ -1 +1,12 @@
|
|
| 1 |
-
LayoutLMv2 fine-tuned on the FUNSD dataset. Code and results are available at the official GitHub repository of my [Master Degree thesis ](https://github.com/AleRosae/thesis-layoutlm)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
LayoutLMv2 fine-tuned on the FUNSD dataset. Code and results are available at the official GitHub repository of my [Master Degree thesis ](https://github.com/AleRosae/thesis-layoutlm).
|
| 2 |
+
|
| 3 |
+
Results obtained with seqeval in strict mode:
|
| 4 |
+
|
| 5 |
+
| | Precision | Recall | F1-score | Variance (F1) |
|
| 6 |
+
|--------------|-----------|--------|----------|---------------|
|
| 7 |
+
| ANSWER | 0.82 | 0.83 | 0.82 | 4e-4 |
|
| 8 |
+
| HEADER | 0.65 | 0.58 | 0.62 | 9e-4 |
|
| 9 |
+
| QUESTION | 0.87 | 0.83 | 0.85 | 3e-5 |
|
| 10 |
+
| Micro avg | 0.84 | 0.82 | 0.82 | 1e-4 |
|
| 11 |
+
| Macro avg | 0.79 | 0.75 | 0.76 | 4e-4 |
|
| 12 |
+
| Weighted avg | 0.84 | 0.82 | 0.82 | 1e-4 |
|