Multimodal Model README

This is a multimodal model for predicting the retention time of molecules under various temperature ramping curves.

Environment Setup

Required environment includes:

PyTorch Version: 2.3.0+cu121
torch_geometric Version: 2.5.3

Library Installation

Use the following commands to install the required libraries:

pip install torch==2.3.0+cu121 -f https://download.pytorch.org/whl/cu121/torch_stable.html
pip install torch_geometric==2.5.3 -f https://data.pyg.org/whl/torch-2.3.0+cu121.html

Running the Code

Set configuration parameters in the ./Model/config.py file.
Run main.py in the root directory of the project:

python main.py

Usage Instructions

Input the SMILES strings of the chemical molecules you are interested in into the predict_mole_smi list (these can be exported from ChemDraw):
```
predict_mole_smi = ['CCCCCCC', 'CCCCCCCCCCCC', 'CCCCCCCCCCCCCCCCCCC']
```
Input the conditions of the temperature ramping curve into the exp_cond list (initial temperature, final temperature, heating rate, initial temperature holding time):
```
exp_cond = [40, 200, 20, 2]
```
Run main.py to predict the retention time of the chemical molecule under the specified temperature ramping curve:
```
python main.py
```

Modify the GNN.GNN_mode parameter to perform training and testing:

Train Mode:

GNN.GNN_mode = 'Train'
data.prepare_gnn_3D()
GNN.num_iterations = 1500
GNN.Mode()

Test Mode:

GNN.GNN_mode = 'Test'
GNN.Test_model_path = '/model_save_600.pth'
GNN.Mode()

Pre Mode:

GNN.GNN_mode = 'Pre'
GNN.predict_model_path = '/model_save_600.pth'
predict_mole_smi = ['CCCCCCC', 'CCCCCCCCCCCC', 'CCCCCCCCCCCCCCCCCCC']
exp_cond = [40, 200, 20, 2]
generate_excel(predict_mole_smi, exp_cond)
GNN.unknown_descriptor_path = './Data/waiting_pre.xlsx'
GNN.Mode()

Notes

Before running the code, ensure all configuration parameters are correctly set in the ./Model/config.py file.
Ensure that the installed versions of PyTorch and torch_geometric meet the requirements to avoid compatibility issues.

By following these steps, you can use this multimodal model to predict the retention time of chemical molecules under specific temperature ramping curves.