Instructions to use GerMedBERT/medbert-512 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use GerMedBERT/medbert-512 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("fill-mask", model="GerMedBERT/medbert-512")# Load model directly from transformers import AutoTokenizer, AutoModelForMaskedLM tokenizer = AutoTokenizer.from_pretrained("GerMedBERT/medbert-512") model = AutoModelForMaskedLM.from_pretrained("GerMedBERT/medbert-512") - Notebooks
- Google Colab
- Kaggle
Passing 'use_auth_token' in the Spacy Config file for NER Training
I am trying to use medbert.de for a fine-tuning NER task. I am aware that medbert.de is a private repository and one needs to use an authentication token to load it.
My question would be, how and where to pass the use_auth_token parameter. The screenshot of the config file is attached below
You could try to use the token by embedding it as an environment variable, or login with the HF cli before. I haven't found anything in the spacy docs.
@thedust Thank you for reply. The second option of logging in with the HF cli should create a problem, as it needs us to pass to use the use_auth_token as False.
Could you be a little more specific about using the token as an environment variable ? Unfortunately, I could not find anything in the spacy docs too. Thanks in advance
