smi-ted 1.0.0

SMILES-based Transformer Encoder-Decoder (SMI-TED)

SMILES-based Transformer Encoder-Decoder (SMI-TED)

This repository is a fork of the original source code and adapted for usage with HuggingFace AutoModel.

Forked GitHub: GitHub Link

Forked HuggingFace: HuggingFace Link

The original repository provides PyTorch source code associated with the publication, "A Large Encoder-Decoder Family of Foundation Models for Chemical Language".

Paper: Arxiv Link

Original GitHub: GitHub Link

Original HuggingFace: HuggingFace Link

For more information contact: [e-mail] or [e-mail].

Usage

import torch
import smi_ted
from transformers import AutoConfig, AutoModel, AutoTokenizer

config = AutoConfig.from_pretrained("bisectgroup/materials-smi-ted-fork")
tokenizer = AutoTokenizer.from_pretrained("bisectgroup/materials-smi-ted-fork")
model = AutoModel.from_pretrained("bisectgroup/materials-smi-ted-fork")

model.smi_ted.tokenizer = tokenizer
model.smi_ted.set_padding_idx_from_tokenizer()

smiles = ['CC1C2CCC(C2)C1CN(CCO)C(=O)c1ccc(Cl)cc1',
'COc1ccc(-c2cc(=O)c3c(O)c(OC)c(OC)cc3o2)cc1O',
'CCOC(=O)c1ncn2c1CN(C)C(=O)c1cc(F)ccc1-2',
'Clc1ccccc1-c1nc(-c2ccncc2)no1',
'CC(C)(Oc1ccc(Cl)cc1)C(=O)OCc1cccc(CO)n1']

with torch.no_grad():
    encoder_outputs = model.encode(smiles)
with torch.no_grad():
    decoded_smiles = model.decode(encoder_outputs)