Skip to content

german medbert model details #8266

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
julien-c merged 2 commits into from
Nov 6, 2020
Merged

german medbert model details #8266

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d138571
model details
smanjil Nov 3, 2020
9de30d0
Apply suggestions from code review
julien-c Nov 6, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
33 changes: 33 additions & 0 deletions model_cards/smanjil/German-MedBERT/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,33 @@
---
language: de
Copy link
Member

@julien-c julien-c Nov 6, 2020

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You can add links to the datasets with

datasets:
- ...

---

# German Medical BERT

This is a fine-tuned model on Medical domain for German language and based on German BERT.

## Overview
**Language model:** bert-base-german-cased

**Language:** German

**Fine-tuning:** Medical articles (diseases, symptoms, therapies, etc..)

**Eval data:** NTS-ICD-10 dataset (Classification)

**Infrastructure:** Gogle Colab


## Details
- We fine-tuned using Pytorch with Huggingface library on Colab GPU.
- With standard parameter settings for fine-tuning as mentioned in original BERT's paper.
- Although had to train for upto 25 epochs for classification.

## Performance (Micro precision, recall and f1 score for multilabel code classification)
![performance](https://raw.githubusercontent.com/smanjil/finetune-lm/master/performance.png)

## Author
Manjil Shrestha: `shresthamanjil21 [at] gmail.com`

Get in touch:
[LinkedIn](https://www.linkedin.com/in/manjil-shrestha-038527b4/)