LTEnjoy commited on
Commit
a63f7d7
1 Parent(s): 6c234b7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -2
README.md CHANGED
@@ -1,7 +1,7 @@
1
  ---
2
  license: mit
3
  ---
4
- Github repo: https://github.com/westlake-repl/ProTrek
5
 
6
  ## Overview
7
  ProTrek is a multimodal model that integrates protein sequence, protein structure, and text information for better
@@ -11,10 +11,12 @@ does.
11
 
12
  ## Model architecture
13
  Protein sequence encoder: [esm2_t12_35M_UR50D](https://huggingface.co/facebook/esm2_t12_35M_UR50D)
 
14
  Protein structure encoder: foldseek_t12_35M (identical architecture with esm2 except that the vocabulary only contains 3Di tokens)
 
15
  Text encoder: [BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext](https://huggingface.co/microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext)
16
 
17
- ## Obtain embeddings and calculate similarity score (please clone the repo first)
18
  ```
19
  import torch
20
 
 
1
  ---
2
  license: mit
3
  ---
4
+ **Github repo: https://github.com/westlake-repl/ProTrek**
5
 
6
  ## Overview
7
  ProTrek is a multimodal model that integrates protein sequence, protein structure, and text information for better
 
11
 
12
  ## Model architecture
13
  Protein sequence encoder: [esm2_t12_35M_UR50D](https://huggingface.co/facebook/esm2_t12_35M_UR50D)
14
+
15
  Protein structure encoder: foldseek_t12_35M (identical architecture with esm2 except that the vocabulary only contains 3Di tokens)
16
+
17
  Text encoder: [BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext](https://huggingface.co/microsoft/BiomedNLP-BiomedBERT-base-uncased-abstract-fulltext)
18
 
19
+ ## Obtain embeddings and calculate similarity score (please clone our repo first)
20
  ```
21
  import torch
22