# Data
All additionl data can be found [here](https://sid.erda.dk/sharelink/EzFLHhjSI3)

## Raw
Contains original data from sources and is split up in:
- literature/ See the folder README.md for more
- out_of_sample/ See the folder README.md for more

## interim
Contains interim results for the `literature` and `out of sample` datasets.
### literature
| Folder/datasets    | Description |
| :------- |:---------|
| `calc_smi2gcs` | .xyz files from xtb calculations for all SMILES.
| `calc_sterics` | steric descriptors all SMILES using MORFEUS.
| `merged` | dataset containing literature data with smi2gcs calculation.
| `smi2gcs_cache` | dataset containing literature data with smi2gcs calculations.


### out_of_sample
| Folder/datasets    | Description |
| :------- |:---------|
| `calc_smi2gcs` | .xyz files from xtb calculations for all SMILES.
| `calc_sterics` | steric descriptors all SMILES using MORFEUS.
| `merged` | dataset containing literature data with smi2gcs calculation.
| `out_of_sample_smi2gcs.json.gz` | dataset containing out of sample dataset with smi2gcs calculations.
| `out_of_sample_sterics.json.gz` | dataset containing out of sample dataset with steric calculations.

## ligands
| Folder/datasets    | Description |
| :------- |:---------|
| `ligand_sterics` | .xyz files from xtb calculations for all SMILES.
| `ligand_xyz` | steric descriptors all SMILES using MORFEUS.
| `merged` | dataset containing literature data with smi2gcs calculation.


## processed
Contains the final datasets used during the paper

| Datasets    | Description |
| :------- |:---------|
| `literature_merged.json.gz` | literature dataset with all features.
| `literature_regio_yield_above_30.json.gz` | literature dataset filtered by top2 reagents and yield above 30 %. Includes all features.
| `lsfml_benchmark/literature_regio_train_test_all_features.json.gz` | dataset that includes the train/test splits for each iteration for lsfml_benchmark.
| `no_duplicate_substrates.json.gz` | the no_duplicate substrate dataset.
| `top2_ir_precatalyst_with_ligands.json.gz` | top2 ir precatalystst dataset with ligand features.
| `top2_ir_precatalyst.json.gz` | top2 ir precatalystst dataset.