Skip to content

Latest commit

 

History

History
37 lines (28 loc) · 1.25 KB

INSTALL.md

File metadata and controls

37 lines (28 loc) · 1.25 KB
Raw

Installation

Requirements

All the codes are tested in the following environment:

  • Linux (tested on Ubuntu 16.04)
  • Python 3.6+
  • PyTorch 1.7.0, PyTorch 1.8.0, PyTorch 1.8.1
  • CUDA 11.1
  • spconv v2.x
  • gcc version >= 5.4.0

Install 3DTrans

NOTE: Please re-install 3DTrans by running python setup.py develop even if you have already installed previous version.

a. Clone this repository.

git clone https://github.com/PJLab-ADG/3DTrans.git

b. Install the dependent libraries as follows:

  • Install the python dependent libraries.

      pip install -r requirements.txt

  • Install the gcc library, we use the gcc-5.4 version

  • Install the SparseConv library, we use the implementation from [spconv].

    • It is recommended that you should install the latest spconv v2.x with pip, see the official documents of spconv.
    • Also, you should choice the right version of spconv, according to your CUDA version. For example, for CUDA 11.1, pip install spconv-cu111

c. Install this pcdet library and its dependent libraries by running the following command:

python setup.py develop