Codemaster67

Add SPE tokenizer (434 SMILES subword tokens) + <|start_of_smiles|>/<|end_of_smiles|> special tokens. Trained on 2M ZINC20 + 2M ChEMBL canonical SMILES. SPE vocab_size=300, min_freq=4000.

9f054a4 verified 6 days ago

raw

history contribute delete

3.65 MB

File too large to display, you can check the raw version instead.