Update README to include disease-centric splits
Browse files
README.md
CHANGED
|
@@ -47,8 +47,9 @@ PROTON is a a 578-million-parameter heterogeneous graph transformer for neurolog
|
|
| 47 |
- `edge_types.pt`: Ordered list of 47 edge types in NeuroKG to create edge type IDs.
|
| 48 |
- `embeddings.pt`: Store of learned embeddings for all 147,020 nodes in NeuroKG (shape `[147020, 512]`).
|
| 49 |
- `embeddings.csv`: Embedding store as a CSV file.
|
|
|
|
| 50 |
|
| 51 |
-
For more details, please refer to our [project website](https://zitniklab.hms.harvard.edu/PROTON).
|
| 52 |
|
| 53 |
|
| 54 |
## Usage Instructions
|
|
|
|
| 47 |
- `edge_types.pt`: Ordered list of 47 edge types in NeuroKG to create edge type IDs.
|
| 48 |
- `embeddings.pt`: Store of learned embeddings for all 147,020 nodes in NeuroKG (shape `[147020, 512]`).
|
| 49 |
- `embeddings.csv`: Embedding store as a CSV file.
|
| 50 |
+
- `disease_splits/`: Directory containing embeddings of PROTON trained on disease-centric splits.
|
| 51 |
|
| 52 |
+
Files within the `disease_splits/` directory follow the naming convention `{node_id}_{artifact}`, where `{node_id}` represents the unique identifier for the disease node in [NeuroKG](https://doi.org/10.7910/DVN/ZDLS3K). For more details, please refer to our [project website](https://zitniklab.hms.harvard.edu/PROTON).
|
| 53 |
|
| 54 |
|
| 55 |
## Usage Instructions
|