Ankh
Ankh copied to clipboard
agemagician
Ankh: Optimized Protein Language Model
Hello, Thanks for your work! I am trying to fine-tune the model on some of my data on a multilabel classification task. From my understanding of the example notebooks, the...
Hi, could you tell me what model you used for the remote homology prediction task please ? I was able to run your example notebooks for secondary structure prediction but...
Hi there! Awesome work! Quick question, in the paper you mention > performed a full demasking/reconstruction of the sequence (i.e., all tokens are reconstructed as individual tokens). Isn't it different...
Hello! I see that this implementation of T5 has ~115 extra ids for masking. Is it possible to tile or pattern these in someway to get the model to generate...
Hello, I have just read the paper, Great work! I hoped to get more insights into the model architecture and training details, but none seems available on the repo. It...
Hi! Are there any publicly available training/fine-tuning scripts? I would like to fine-tune the pre-trained Ankh on a custom dataset of protein sequences using the masked setup described in section...
Hi there, great work on making these large pre-trained protein LMs available for us to use. Are the fine-tuned models for prediction of secondary structure, contacts, solubility, fluorescence etc going...
Hi, thanks so much for your work ! I was wondering, using Ankh, have you generated the embeddings on uniref50 or another database and made it available somewhere by any...
Hello, I encountered an issue while trying to use the ankh2-large model on the Hugging Face model hub. Here is the code snippet I used: ```bash from transformers import AutoTokenizer,...
