how to input Anndata by scanpy
#400
by
yihang0017
- opened
"I hope it can support directly inputting .h5ad data and automatically completing preprocessing. Otherwise, the usability of Geneformer will be greatly reduced for others. Also, where can the preprocessing code for the paper be found?"
Thank you for your question!
Please see the tokenizer documentation, including the option for using anndata: https://geneformer.readthedocs.io/en/latest/geneformer.tokenizer.html
The preprocessing code is in the tokenizer:
https://huggingface.co/ctheodoris/Geneformer/blob/main/geneformer/tokenizer.py
ctheodoris
changed discussion status to
closed