基于分子/离子构型的晶体结构设计软件通过结合经验公式、机器学习势函数微调、第一性原理分步优化和分子/离子识别技术,实现了从分子/离子构型出发的高效晶体结构设计。该软件采用模块化设计,支持全流程自动化材料筛选,在保证预测精度的同时显著提升计算效率。
- 双模块工作流
- EE模块:基于经验评估的离子组合生成
- CSP模块:基于离子晶体结构预测的优化筛选
- 智能任务管理
- 实时进程监控(PID跟踪)
- 日志文件自动符号链接
- 进程安全终止与资源清理
- 高级日志系统
- 分页浏览(10条/页)
- 模块过滤(CSP/EE)
- 软链接解析显示实际路径
- 跨平台支持(Linux/Docker)
- 基于Python 3.11+的面向对象架构
- 集成psutil进程管理
- 结构化日志记录系统
组件 | 最低版本 |
---|---|
Python | 3.11 |
psutil | 5.9 |
Docker | 20.10 |
ase | 3.23.0 |
deepmd-kit | 3.0.1 |
torch | 2.5.0 |
dpdispatcher | 0.6.7 |
numpy | 1.26.4 |
paramiko | 3.5.1 |
pyxtal | 1.0.4 |
phonopy | 2.28.0 |
rdkit | 2024.03.3 |
pip install ion-csp
或
# 创建虚拟环境
python -m venv venv
source venv/bin/activate # Linux/Mac
# 安装依赖
git clone https://github.com/Bagabaga007/ion_CSP.git
cd ion_CSP
pip install -e .
ion-csp
启动交互式命令行界面,支持以下操作:
- 模块选择
- 日志查看
- 进程管理
./scripts/main_EE.sh examples/example_1
从SMILES表格生成离子组合
./scripts/main_CSP.sh examples/example_2
从离子组合生成并优化晶体结构
graph TD
A[用户界面] --> B[任务管理器]
B --> C[EE模块]
B --> D[CSP模块]
B --> E[日志系统]
B --> F[任务调度]
C --> G[经验评估引擎]
D --> H[离子晶体结构预测引擎]
E --> I[结构化日志]
F --> J[进程终止]
- Fork仓库并创建特性分支
- 编写单元测试覆盖新功能
- 提交Pull Request时注明关联Issue
- 遵循PEP8代码规范
本项目采用MIT许可证,详见LICENSE文件。
This software enables efficient crystal structure design from molecular/ion configurations by integrating empirical formulas, tuned machine learning potentials, stepwise first-principles optimization, and molecular/ion recognition techniques. The modular architecture ensures extensibility and maintainability while maintaining prediction accuracy.
- Dual-Module Workflow
- EE Module: Empirical evaluation-based ion combination generation
- CSP Module: Ion crystal structure prediction and optimization
- Intelligent Task Management
- Real-time process monitoring (PID tracking)
- Automatic log file symlink creation
- Safe process termination with resource cleanup
- Advanced Logging System
- Paginated log viewing (10 entries/page)
- Module-based filtering (CSP/EE)
- Symlink resolution for actual log paths
- Cross-platform support (Linux/Docker)
- Object-oriented architecture with Python 3.11+
- Integrated process management via psutil
- Structured logging system
Component | Min Version |
---|---|
Python | 3.11 |
psutil | 5.9 |
Docker | 20.10 |
ase | 3.23.0 |
deepmd-kit | 3.0.1 |
torch | 2.5.0 |
dpdispatcher | 0.6.7 |
numpy | 1.26.4 |
paramiko | 3.5.1 |
pyxtal | 1.0.4 |
phonopy | 2.28.0 |
rdkit | 2024.03.3 |
# Create virtual environment
python -m venv venv
source venv/bin/activate # Linux/Mac
# Install dependencies
git clone https://github.com/Bagabaga007/ion_CSP.git
cd ion_CSP
pip install -e .
ion-csp
Launches CLI interface with:
- Module selection
- Log management
- Process control
./scripts/main_EE.sh examples/example_1
Generates ion combinations from SMILES tables
./scripts/main_CSP.sh examples/example_2
Optimizes crystal structures from ion combinations
graph TD
A[User Interface] --> B[Task Manager]
B --> C[EE Module]
B --> D[CSP Module]
B --> E[Log System]
B --> F[Task Scheduler]
C --> G[Empirical Evaluation Engine]
D --> H[Ionic Crystal Structure Prediction Engine]
E --> I[Structured Logs]
F --> J[Process Termination]
- Fork repository and create feature branch
- Write unit tests for new features
- Submit PR with issue reference
- Follow PEP8 coding standards
MIT License, see LICENSE file.