RPSubAlign

This repository contains OpenNMT-py trained models for RPSubAlign.

RPSA_selfies.pt: Model trained using SELFIES.
RPSA_smiles.pt: Model trained using SMILES.

Usage

To load the model, you can download it from Hugging Face:

wget https://huggingface.co/Aminoacid1226/RPSA/resolve/main/RPSA_smiles.pt

Or manually download the model from the "Files and versions" page.

Model Usage

For detailed usage instructions, please refer to the OpenNMT manual: OpenNMT-py Translate Options.

After downloading the model and preparing the configuration file, you can use the following command for prediction:

onmt_translate -config translate.yml

Configuration File

The configuration file (translate.yml) should include at least the following parameters:

model: <path_to_model>  # Path to the model file
src: <path_to_source_file>  # Path to the input molecular file
output: <path_to_output_file>  # Path to save prediction results

Additional Parameters

gpu: Specify the GPU device ID (if using GPU)
beam_size: Set the beam search size
n_best: Number of returned predictions

For more available parameters, please refer to the OpenNMT manual.

train from scratch

If you want to train your own model from scratch, check out our GitHub home page