search

DeepGraphLearning / ESM-GearNet

ESM-GearNet for Protein Structure Representation Learning (https://arxiv.org/abs/2303.06275)
79 stars 10 forks source link

Use pretrained ESM-GearNet model to acquire sequence embeddings, like ESM #2

Open deardeer7 opened 8 months ago

deardeer7 commented 8 months ago

Great work! I am attemptting to use your pretrained ESM-GearNet model to acquire sequence embeddings, similar to the ones from ESM. How can I achieve this?

ShallowDream0218 commented 7 months ago

Great work! I am attemptting to use your pretrained ESM-GearNet model to acquire sequence embeddings, similar to the ones from ESM. How can I achieve this?

Great work! I am attemptting to use your pretrained ESM-GearNet model to acquire sequence embeddings, similar to the ones from ESM. How can I achieve this?

I want to extract sequence embeddings by using this model . Have you achieved yet?

Yangqy-16 commented 7 months ago

Same question. It seems that the authors of this work only gave the commands for training from scratch but didn't release the pretrained model. So how can we obtain sequence embeddings from the model without training by ourselves? Seeks for help.

ShallowDream0218 commented 7 months ago

Same question. It seems that the authors of this work only gave the commands for training from scratch but didn't release the pretrained model. So how can we obtain sequence embeddings from the model without training by ourselves? Seeks for help.

What's the result of training from scratch? Isn't it sequence embeddings?

Yangqy-16 commented 7 months ago

Same question. It seems that the authors of this work only gave the commands for training from scratch but didn't release the pretrained model. So how can we obtain sequence embeddings from the model without training by ourselves? Seeks for help.

What's the result of training from scratch? Isn't it sequence embeddings?

Sorry, I don't have the resource (4 A100 GPUs as mentioned in the article) to train from scratch so I don't know the result.