Skip to content

GitLab

Commit 07104324 authored by dengjb's avatar dengjb
Browse files

Update README.md

parent e38b4baa
Hide whitespace changes
Inline Side-by-side
Showing with 36 additions and 3 deletions
README.md
View file @ 07104324
...@@ -50,10 +50,37 @@ deepspeed: 0.14.2+das.opt2.dtk2504 ...@@ -50,10 +50,37 @@ deepspeed: 0.14.2+das.opt2.dtk2504
其它非深度学习库安装方式如下： 其它非深度学习库安装方式如下：
```bash ```bash
cd /your_code_path/pointpillars-pytorch
pip install -r requirements.txt pip install -r requirements.txt
```
安装pointpillars
```bash
cd /your_code_path/pointpillars-pytorch
python setup.py build_ext --inplace
pip install .
``` ```
## 数据集 ## 数据集
下载数据集[point cloud](https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_velodyne.zip)(29GB), [images](https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_image_2.zip)(12 GB), [calibration files](https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_calib.zip)(16 MB)和[labels](https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_label_2.zip)(5 MB)。数据格式如下所示: 1、下载数据集[point cloud](https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_velodyne.zip)(29GB), [images](https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_image_2.zip)(12 GB), [calibration files](https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_calib.zip)(16 MB)和[labels](https://s3.eu-central-1.amazonaws.com/avg-kitti/data_object_label_2.zip)(5 MB)。
解压后数据格式如下所示:
```
kitti
|- training
|- calib (#7481 .txt)
|- image_2 (#7481 .png)
|- label_2 (#7481 .txt)
|- velodyne (#7481 .bin)
|- testing
|- calib (#7518 .txt)
|- image_2 (#7518 .png)
|- velodyne (#7518 .bin)
```
2、制作数据集：
```bash
cd /your_code_path/pointpillars-pytorch
python pre_process_kitti.py --data_root your_path_to_kitti
```
制作完数据结构如下所示
``` ```
kitti kitti
|- training |- training
...@@ -74,15 +101,16 @@ pip install -r requirements.txt ...@@ -74,15 +101,16 @@ pip install -r requirements.txt
|- kitti_infos_test.pkl |- kitti_infos_test.pkl
|- kitti_dbinfos_train.pkl |- kitti_dbinfos_train.pkl
``` ```
## 训练 ## 训练
``` ```
cd PointPillars/ cd /your_code_path/pointpillars-pytorch
python train.py --data_root your_path_to_kitti python train.py --data_root your_path_to_kitti
``` ```
## 推理-验证 ## 推理-验证
``` ```
cd PointPillars/ cd /your_code_path/pointpillars-pytorch
python evaluate.py --ckpt pretrained/epoch_160.pth --data_root your_path_to_kitti python evaluate.py --ckpt pretrained/epoch_160.pth --data_root your_path_to_kitti
``` ```
...@@ -117,6 +145,11 @@ AOS AP: 74.4584 68.3366 65.2222 ...@@ -117,6 +145,11 @@ AOS AP: 74.4584 68.3366 65.2222
bbox_bev AP: 78.1073 70.4609 66.9258 bbox_bev AP: 78.1073 70.4609 66.9258
bbox_3d AP: 72.8538 62.0781 59.3903 bbox_3d AP: 72.8538 62.0781 59.3903
``` ```
## 环境配置信息
DCU型号：K100AI
节点数量：1
卡数：1
## 应用场景 ## 应用场景
### 算法类别 ### 算法类别
文本理解 文本理解
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment