# PCT-Cap

**Repository Path**: oscc-project/pct-cap

## Basic Information

- **Project Name**: PCT-Cap
- **Description**: No description available
- **Primary Language**: Python
- **License**: MulanPSL-2.0
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2024-06-14
- **Last Updated**: 2024-08-29

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# PC-Cap #
This repository holds the main resources for the Point Cloud for Capacitance (PC-Cap) project.

This project delivers an idea of representing post-layout data by point cloud. 

This project is an instance of AI for 3D pattern matching in the area of capacitance extraction.

![](/imgs/flow_3d_pattern_matching.png "The flow of 3D capacitance extraction based on neural network")

## News ## 
- 2024.05.12: Presentation in [ISEDA Technical Session 10: Advanced in EMIR and Parasitic Extraction](https://www.eda2.com/iseda/session10.html)
- 2024.03.27: [Related paper](https://ieeexplore.ieee.org/document/10617673) was accepted by 2024 International Symposium of Electronics Design Automation (ISEDA)

## Dataset ##

The PC-Cap project has built two datasets: [3D dataset](https://gitee.com/oscc-project/i-bm/tree/master/PC-Cap_3D_Dataset) and  [Width-Etch dataset](https://gitee.com/oscc-project/i-bm/tree/master/PC-Cap_WE_Dataset).

For more details on the dataset please visit [iEDA-iBM](https://gitee.com/oscc-project/i-bm).

## Quick Start ##

Our OS information is **Ubuntu 20.04.5 LTS**.

1. install the environment if necessary
```shell
conda create --name <env> --file environment.yml
conda activate <env>
```

2. check the integrity of dataset 
```shell
python utils/check_pc_path_available.py --ann <path/to/pc_annotation.csv>
```

3. run the train.py
```shell
python train.py --train <path/to/train.csv> --test <path/to/test.csv>
```

4. some further data analyses are shown in **utils/utils.py**  
Such as dataset distribution, model accuracy, algorithm visualization, etc.

## Citation ## 
If you benefit from our work in your research, please consider to cite the following papers:
```txt
@INPROCEEDINGS{PCT-Cap,
  author={Cai, Ye and Liang, Yuyao and Luo, Zhipeng and Xie, Biwei and Li, Xingquan},
  booktitle={2024 2nd International Symposium of Electronics Design Automation (ISEDA)}, 
  title={PCT-Cap: Point Cloud Transformer for Accurate 3D Capacitance Extraction}, 
  year={2024},
  pages={421-426},
  doi={10.1109/ISEDA62518.2024.10617673}
}

```

## Thanks ##
| Sources | Descriptions |
| --- | --- |
| [yanx27/Pointnet_Pointnet2_pytorch](https://github.com/yanx27/Pointnet_Pointnet2_pytorch) | PointNet2Cap was based on PointNet2 with modifications on the head |
| [MenghaoGuo/PCT](https://github.com/MenghaoGuo/PCT) | PCT implementation |
| [d2l-ai/d2l-zh](https://github.com/d2l-ai/d2l-zh) | [ResNet implementation](https://zh.d2l.ai/chapter_convolutional-modern/resnet.html) |
---



---
Please feel free to contact us if you have any questions.

Email: