Instructions to use NitzanBar/umls-spanbert with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use NitzanBar/umls-spanbert with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="NitzanBar/umls-spanbert")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("NitzanBar/umls-spanbert") model = AutoModelForSequenceClassification.from_pretrained("NitzanBar/umls-spanbert") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,13 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
Based ob the paper: UmlsBERT: Augmenting Contextual Embeddings with a Clinical Metathesaurus
|
| 5 |
+
and the github repo: https://github.com/gmichalo/UmlsBERT
|
| 6 |
+
|
| 7 |
+
Changing base model to SpanBert instead of Bert.
|
| 8 |
+
Trained from scratch on MIMIC dataset, using the UMLS dataset to mask words within the text.
|
| 9 |
+
|
| 10 |
+
We achived better accuracy on MedNLI dataset.
|
| 11 |
+
|
| 12 |
+
Bert Model accuracy: 83%
|
| 13 |
+
SpanBert Model accuracy: 86%
|