copuladock / code /compact_v1 /README.md
liofoil's picture
Add files using upload-large-folder tool
0cb481f verified
|
Raw
History Blame Contribute Delete
1.35 kB
# compact_v1 construction and reader code
This directory contains the minimum code needed to read the released compact shards and to rebuild the same format from PDB docking poses.
## Read an existing release
Only `compact_graph_dataset.py` is required for normal training. It provides `CompactGraphDataset`, which opens tensor-only shard files lazily with `torch.load(..., mmap=True)` and reconstructs standard PyTorch Geometric `Data` objects.
## Build compact shards from PDB poses
Keep these three files in the same directory because the builder imports the other two by filename:
- `build_compact_v1_direct.py`
- `build_graph_unified_enhanced.py`
- `convert_to_compact_v1.py`
The direct builder expects a flat directory with one subdirectory per system containing `protein.pdb`, `ligand.pdb`, and predicted pose `.pdb` files. `materialize_hiqbind_gnncp.py` converts Docking Base common outputs to that layout through hard links.
`build_compact_v1_direct.sbatch` is an example Slurm wrapper only. Review and adapt account, partition, Python environment, paths, CPU count, and memory for your own cluster before use.
Run the bundled unit tests after changing the format code:
```bash
python -m unittest -v test_compact_graph_dataset.py test_build_compact_v1_direct.py
```
See `../../docs/DATASET_USAGE_ZH.md` for the Chinese user guide.