AttributeError: 'ESMProteinError' object has no attribute 'hidden_states'

[jinlong-abc]

---

AttributeError Traceback (most recent call last) Cell In[58], line 5 1 import torch 3 # we'll summarize the embeddings using their mean across the sequence dimension 4 # which allows us to compare embeddings for sequences of different lengths ----> 5 all_mean_embeddings = [ 6 torch.mean(output.hidden_states, dim=-2).squeeze() for output in outputs 7 ] 9 # now we have a list of tensors of [num_layers, hidden_size] 10 print("embedding shape [num_layers, hidden_size]:", all_mean_embeddings[0].shape)

Cell In[58], line 6 1 import torch 3 # we'll summarize the embeddings using their mean across the sequence dimension 4 # which allows us to compare embeddings for sequences of different lengths 5 all_mean_embeddings = [ ----> 6 torch.mean(output.hidden_states, dim=-2).squeeze() for output in outputs 7 ] 9 # now we have a list of tensors of [num_layers, hidden_size] 10 print("embedding shape [num_layers, hidden_size]:", all_mean_embeddings[0].shape)

AttributeError: 'ESMProteinError' object has no attribute 'hidden_states'

[SakshiTaware]

I had the same error, any update on how to solve it?

[jinlong-abc]

这是来自QQ邮箱的假期自动回复邮件。   您好，我最近正在休假中，无法亲自回复您的邮件。我将在假期结束后，尽快给您回复。

[SHG369]

I met the same error, can it be solved now?

[Sparadox001]

same problem

AttributeError Traceback (most recent call last) Cell In[14], line 6 1 import torch 3 # we'll summarize the embeddings using their mean across the sequence dimension 4 # which allows us to compare embeddings for sequences of different lengths 5 all_mean_embeddings = [ ----> 6 torch.mean(output.hidden_states, dim=-2).squeeze() for output in outputs 7 ] 9 # now we have a list of tensors of [num_layers, hidden_size] 10 print("embedding shape [num_layers, hidden_size]:", all_mean_embeddings[0].shape)

AttributeError: 'ESMProteinError' object has no attribute 'hidden_states'

[HansJWolters]

Same here