A Jittor implementation of Point Cloud Transformer (PCT) for ModelNet40 classification.
Introduction
This project implements a point cloud classification model using the Jittor deep learning framework. The model is based on Point Cloud Transformer (PCT) and is used for the ModelNet40 classification task.
The program reads preprocessed point cloud data from .npy files, trains a neural network with data augmentation, and generates predictions for the test set in result.json.
PCT_jittor
A Jittor implementation of Point Cloud Transformer (PCT) for ModelNet40 classification.
Introduction
This project implements a point cloud classification model using the Jittor deep learning framework. The model is based on Point Cloud Transformer (PCT) and is used for the ModelNet40 classification task.
The program reads preprocessed point cloud data from
.npyfiles, trains a neural network with data augmentation, and generates predictions for the test set inresult.json.Framework
Project Structure
```text PCT_jittor/ ├── pct.py ├── README.md └── .gitignore