# SSMVCL

**Repository Path**: Tomhappy/SSMVCL

## Basic Information

- **Project Name**: SSMVCL
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2026-02-01
- **Last Updated**: 2026-02-01

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# SSMVCL

## Environment Settings
> python==3.10.14 \
> scipy==1.13.0 \
> torch==2.0.1 \
> numpy==1.26.0 \
> scikit_learn==1.5.2

GPU: NVIDIA RTX 4050 GPU \
CPU: Intel Core i5-13500H CPU

## Data preprocessing
1. Run neibor.py to get the structure view of neighbors.
2. Run mp_gen.py to get the semantic view metapath node matrix.
3. Run pos.py to obtain positive and negative samples for view contrast learning.
4. Run split_data_kfold.py to perform n-fold splitting of the dataset.

## Model
1. SSMVCL.py is the main model content.
2. mp_encoder.py is semantic-aware view encoding.
3. sc_encoder.py is structure-aware view encoding.
4. contrast.py is view contrast learning.
5. params.py contains detailed parameter information for each dataset and task.

## Usage
1. You need to process the data.
2. Run main.py.