jspector792
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -4,4 +4,47 @@ tags:
|
|
4 |
- biology
|
5 |
size_categories:
|
6 |
- 100B<n<1T
|
7 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
- biology
|
5 |
size_categories:
|
6 |
- 100B<n<1T
|
7 |
+
---
|
8 |
+
# Aggregated Weight Matrices Used in *Transformers Enhance the Predictive Power of Network Medicine*
|
9 |
+
*Jonah Spector, Andrés Aldana Gonzalez, Michael Sebek, Joseph Ehlert, Christian DeFrondeville, Albert-László Barabási*
|
10 |
+
## Data Structure
|
11 |
+
This directory contains all aggregated weight matrices used to make figures in the manuscript. The data structure is as follows
|
12 |
+
```
|
13 |
+
aggregated_matrices
|
14 |
+
aggregated_attentions
|
15 |
+
dcm_samples
|
16 |
+
fine_tuned
|
17 |
+
max
|
18 |
+
layer_0
|
19 |
+
layer_1
|
20 |
+
layer_2
|
21 |
+
layer_3
|
22 |
+
layer_4
|
23 |
+
layer_5
|
24 |
+
pretrained
|
25 |
+
max
|
26 |
+
genecorpus
|
27 |
+
fine_tuned
|
28 |
+
max
|
29 |
+
pretrained
|
30 |
+
max
|
31 |
+
aggregated_embeddings
|
32 |
+
dcm_samples
|
33 |
+
fine_tuned
|
34 |
+
max
|
35 |
+
layer_input
|
36 |
+
layer_0
|
37 |
+
layer_1
|
38 |
+
layer_2
|
39 |
+
layer_3
|
40 |
+
layer_4
|
41 |
+
layer_5
|
42 |
+
pretrained
|
43 |
+
max
|
44 |
+
genecorpus
|
45 |
+
fine_tuned
|
46 |
+
max
|
47 |
+
pretrained
|
48 |
+
max
|
49 |
+
```
|
50 |
+
In the folder at the end of each branch there are 3 files, called ```counts.pkl```, ```vals.pkl```, ```gene_dict.pkl```. The counts matrix is a square matrix of size NxN, where N is the number of unique genes in the sample set. It counts how many times each pair of genes co-occurs in a sample. The vals matrix is the same shape, and contains the aggregated weight between each pair of genes. ```gene_dict.pkl``` is a dictionary that specifies the index of each gene in the matrix.
|