add Dockerfile

README.md

@@ -19,13 +19,15 @@ Supported backbones:
 - [x] ERFNet
 - [x] VGG
 - [x] MobileNet
+- [] DLA(coming soon)
 
 Supported detectors:
 - [x] [SCNN](configs/scnn)
 - [x] [UFLD](configs/ufld)
 - [x] [RESA](configs/resa)
 - [x] [LaneATT](configs/laneatt)
 - [x] [CondLane](configs/condlane)
+- [] CLRNet(coming soon)
 
 
 ## Installation
@@ -51,10 +53,10 @@ conda activate lanedet
 ```Shell
 # Install pytorch firstly, the cudatoolkit version should be same in your system.
 
-conda install pytorch torchvision cudatoolkit=10.1 -c pytorch
+conda install pytorch==1.8.0 torchvision==0.9.0 cudatoolkit=10.1 -c pytorch
 
 # Or you can install via pip
-pip install torch torchvision
+pip install torch==1.8.0 torchvision==0.9.0
 
 # Install python packages
 python setup.py build develop
@@ -67,7 +69,7 @@ python setup.py build develop
 Download [CULane](https://xingangpan.github.io/projects/CULane.html). Then extract them to `$CULANEROOT`. Create link to `data` directory.
 
 ```Shell
-cd $RESA_ROOT
+cd $LANEDET_ROOT
 mkdir -p data
 ln -s $CULANEROOT data/CULane
 ```
@@ -83,7 +85,7 @@ $CULANEROOT/list                 # data lists
 Download [Tusimple](https://github.com/TuSimple/tusimple-benchmark/issues/3). Then extract them to `$TUSIMPLEROOT`. Create link to `data` directory.
 
 ```Shell
-cd $RESA_ROOT
+cd $LANEDET_ROOT
 mkdir -p data
 ln -s $TUSIMPLEROOT data/tusimple
 ```
@@ -116,7 +118,7 @@ python main.py [configs/path_to_your_config] --gpus [gpu_ids]
 
 For example, run
 ```Shell
-python main.py configs/resa/resa50_culane.py --gpus 0 1 2 3
+python main.py configs/resa/resa50_culane.py --gpus 0
 ```
 
 ### Testing
@@ -127,7 +129,7 @@ python main.py [configs/path_to_your_config] --validate --load_from [path_to_you
 
 For example, run
 ```Shell
-python main.py configs/resa/resa50_culane.py --validate --load_from culane_resnet50.pth --gpus 0 1 2 3
+python main.py configs/resa/resa50_culane.py --validate --load_from culane_resnet50.pth --gpus 0
 ```
 
 Currently, this code can output the visualization result when testing, just add `--view`.

docker/Dockerfile

@@ -0,0 +1,18 @@
+FROM nvcr.io/nvidia/pytorch:21.10-py3
+
+ENV DEBIAN_FRONTEND noninteractive
+
+RUN rm -rf /var/lib/apt/lists/* &&  rm -rf /etc/apt/sources.list.d/*\
+    /etc/apt/sources.list.d/cuda.list \
+    /etc/apt/sources.list.d/nvidia-ml.list && \
+    sed -i s@/archive.ubuntu.com/@/mirrors.ustc.edu.cn/@g /etc/apt/sources.list && \
+    apt-get update && apt-get install -y ffmpeg libsm6 libxext6 git ninja-build libglib2.0-0 libsm6 libxrender-dev libxext6 \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+# Install LaneDet 
+RUN conda clean --all
+RUN git clone https://github.com/turoad/lanedet.git /lanedet
+WORKDIR /lanedet
+RUN pip install --no-cache-dir -r requirements.txt -i https://mirrors.aliyun.com/pypi/simple/ && \
+  pip install --no-cache-dir -e .

docker/README.md

@@ -0,0 +1,22 @@
+# Building a LaneDet container
+
+### Install Docker and NVIDIA Container Toolkit
+
+https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html
+
+### Build Container
+
+> 
+
+```Shell
+# From root of LaneDet repo
+cd $LANEDET_ROOT
+
+# Build:
+docker build -f docker/Dockerfile -t lanedet:latest .
+
+# Run:
+docker run --gpus all -it \
+	--shm-size=8gb  \
+	--name=lanedet --ipc=host --net=host lanedet:latest
+```

requirements.txt

@@ -1,5 +1,5 @@
-torch==1.6.0
-torchvision==0.7.0
+torch==1.8.0
+torchvision==0.9.0
 pandas
 addict
 sklearn
@@ -13,5 +13,5 @@ Shapely==1.7.0
 ujson==1.35
 yapf
 albumentations==0.4.6
-mmcv==0.5.6
+mmcv==1.2.5
 pathspec