|
--- |
|
annotations_creators: [] |
|
language: [] |
|
language_creators: |
|
- expert-generated |
|
license: |
|
- apache-2.0 |
|
multilinguality: |
|
- monolingual |
|
pretty_name: canonicalized PubChem-10m |
|
size_categories: |
|
- 100K<n<1M |
|
source_datasets: |
|
- original |
|
tags: |
|
- PubChem |
|
- chemical |
|
- SMILES |
|
task_categories: [] |
|
task_ids: [] |
|
--- |
|
|
|
### dataset description |
|
We downloaded PubChem-10m dataset from [here](https://deepchemdata.s3-us-west-1.amazonaws.com/datasets/pubchem_10m.txt.zip) and canonicalized it. |
|
We used the following function to canonicalize the data and removed some SMILES that cannot be read by RDKit. |
|
|
|
```python: |
|
from rdkit import Chem |
|
def canonicalize(mol): |
|
mol = Chem.MolToSmiles(Chem.MolFromSmiles(mol),True) |
|
return mol |
|
``` |
|
|
|
We randomly split the preprocessed data into train and validation. The ratio is 9 : 1. |
