Commit ·
6805f21
1
Parent(s): 88fd15a
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,16 +1,25 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
| 4 |
|
| 5 |
-
KEPTlongfomer
|
|
|
|
|
|
|
| 6 |
|
|
|
|
|
|
|
| 7 |
|
| 8 |
-
|
| 9 |
-
|
| 10 |
-
And then pretrained with Hierarchical Self-Alignment Pretrainumls (HSAP) using Knowledge Graph UMLS.
|
| 11 |
This includes (a) Hierarchy, (b) Synonym, (c) Abbreviation. For more info, see section 3.3 in [paper](https://arxiv.org/pdf/2210.03304.pdf).
|
|
|
|
| 12 |
|
| 13 |
-
|
|
|
|
| 14 |
|
| 15 |
With the following result:
|
| 16 |
| Metric | Score |
|
|
@@ -42,5 +51,4 @@ With the following result:
|
|
| 42 |
|acc_macro| =0.08639139221100567|
|
| 43 |
|
| 44 |
|
| 45 |
-
A sister model is available [here](https://huggingface.co/whaleloops/KEPTlongformer-PMM3/).
|
| 46 |
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
+
language: "en"
|
| 4 |
+
tags:
|
| 5 |
+
- longformer
|
| 6 |
+
- clinical
|
| 7 |
+
- biomedical
|
| 8 |
---
|
| 9 |
|
| 10 |
+
<span style="font-size:larger;">**KEPTlongfomer**</span> is a medical knowledge enhanced version of Longformer that was further pre-trained using [contrastive learning](https://arxiv.org/pdf/2210.03304.pdf).
|
| 11 |
+
The model achieves SOTA performance on auto ICD coding on MIMIC-III as of 11/12/2022.
|
| 12 |
+
A sister model for better performance is available [here](https://huggingface.co/whaleloops/KEPTlongformer-PMM3/).
|
| 13 |
|
| 14 |
+
### Pre-training
|
| 15 |
+
We initialized this model from [clinical longformer](https://huggingface.co/yikuan8/Clinical-Longformer).
|
| 16 |
|
| 17 |
+
And then pretrained with Hierarchical Self-Alignment Pretrain (HSAP) using Knowledge Graph UMLS.
|
|
|
|
|
|
|
| 18 |
This includes (a) Hierarchy, (b) Synonym, (c) Abbreviation. For more info, see section 3.3 in [paper](https://arxiv.org/pdf/2210.03304.pdf).
|
| 19 |
+
The learning rate was 5e-5, weight decay was 0.01, adam epsilon was 1e-5.
|
| 20 |
|
| 21 |
+
### Usage
|
| 22 |
+
See our [github](https://github.com/whaleloops/KEPT/tree/rerank300) for how to use this with prompts on auto ICD coding.
|
| 23 |
|
| 24 |
With the following result:
|
| 25 |
| Metric | Score |
|
|
|
|
| 51 |
|acc_macro| =0.08639139221100567|
|
| 52 |
|
| 53 |
|
|
|
|
| 54 |
|