SDLF: Simple Deep Learning Framework

A simple and generalized framework for deep learning projects.

News

2021-02-22: SDLF v1.2.1 released!

2021-02-22: SDLF v1.2.0 released!

2021-02-18: SDLF v1.1.0 released!

2021-02-01: SDLF v1.0.1 released!

2020-12-22: SDLF v1.0 released!

2020-12-16: SDLF v0.9 released!

Install on Ubuntu 16.04 / 18.04

1. Clone this repository

git clone https://github.com/wei-ForPrivateUse/sdlf.git
cd ./sdlf

2. Install dependencies

It is recommended to use Anaconda package manager.

conda install numpy tqdm yaml
conda install -c conda-forge tensorboardX

3. Install Pytorch

This project is tested with pytorch v1.2.0, please follow the instructions in Pytorch.

4. Build and setup this package

Build the package with bdist_wheel (do not use python setup.py install).

python setup.py bdist_wheel
cd ./dist

Finally, use pip to install the generated whl file.

Third-party libraries

This project uses torchplus, which is a part of SECOND.

• SECOND: A 3D object detection algorithm.

License

This project is licensed under the MIT License - see the LICENSE file for details