Update docs

shinya7y · Sep 8, 2021 · c622acf · c622acf
1 parent 4cfd0ac
commit c622acf
Show file tree

Hide file tree

Showing 5 changed files with 155 additions and 8 deletions.
diff --git a/README.md b/README.md
@@ -9,9 +9,9 @@ https://arxiv.org/abs/2103.14027
 
 ## Changelog
 
-- recent:
+- 21.09 (Sept. 2021):
   - Support gradient accumulation to simulate large batch size with few GPUs ([example](configs/manga109/universenet50_2008_fp16_1x4x4_mstrain_480_960_1x_manga109s.py))
-  - Add codes for PVT, PVTv2, DDOD
+  - Add codes for CBNetV2, PVT, PVTv2, DDOD
   - Update and fix codes for mmdet 2.14.0, mmcv-full 1.3.9
 - 21.04 (Apr. 2021):
   - Propose [Universal-Scale object detection Benchmark (USB)](https://arxiv.org/abs/2103.14027)
@@ -39,6 +39,7 @@ https://arxiv.org/abs/2103.14027
 Methods and architectures:
 
 - [x] [UniverseNets (arXiv 2021)](configs/universenet/)
+- [x] [CBNetV2 (arXiv 2021)](configs/cbnet/)
 - [x] [PVT (ICCV 2021), PVTv2 (arXiv 2021)](configs/pvt/)
 - [x] [Swin Transformer (ICCV 2021)](configs/swin/)
 - [x] [DDOD (ACMMM 2021)](configs/ddod/)
@@ -55,6 +56,40 @@ Benchmarks and datasets:
 - [x] [Manga109-s dataset (MTAP 2017, IEEE MultiMedia 2020)](configs/manga109/)
 - [x] [NightOwls dataset (ACCV 2018)](configs/nightowls/)
 
+## Usage
+
+### Installation
+
+See [get_started.md](docs/get_started.md).
+
+### Basic Usage
+
+See [MMDetection documents](#getting-started).
+Especially, see [this document](docs/1_exist_data_model.md) to evaluate and train existing models on COCO.
+
+### Examples
+
+We show examples to evaluate and train UniverseNet-20.08 on COCO with 4 GPUs.
+
+```bash
+# evaluate pre-trained model
+mkdir -p ${HOME}/data/checkpoints/
+wget -P ${HOME}/data/checkpoints/ https://github.com/shinya7y/UniverseNet/releases/download/20.08/universenet50_2008_fp16_4x4_mstrain_480_960_2x_coco_20200815_epoch_24-81356447.pth
+CONFIG_FILE=configs/universenet/universenet50_2008_fp16_4x4_mstrain_480_960_2x_coco.py
+CHECKPOINT_FILE=${HOME}/data/checkpoints/universenet50_2008_fp16_4x4_mstrain_480_960_2x_coco_20200815_epoch_24-81356447.pth
+GPU_NUM=4
+bash tools/dist_test.sh ${CONFIG_FILE} ${CHECKPOINT_FILE} ${GPU_NUM} --eval bbox
+
+# train model
+CONFIG_FILE=configs/universenet/universenet50_2008_fp16_4x4_mstrain_480_960_2x_coco.py
+CONFIG_NAME=$(basename ${CONFIG_FILE} .py)
+WORK_DIR="${HOME}/logs/coco/${CONFIG_NAME}_`date +%Y%m%d_%H%M%S`"
+GPU_NUM=4
+bash tools/dist_train.sh ${CONFIG_FILE} ${GPU_NUM} --work-dir ${WORK_DIR} --seed 0
+```
+
+Even if you have one GPU, we recommend using `tools/dist_train.sh` and `tools/dist_test.sh` to avoid [a SyncBN issue](https://github.com/open-mmlab/mmdetection/issues/847#issuecomment-504421173).
+
 ## Citation
 
 ```

diff --git a/docker/Dockerfile b/docker/Dockerfile
@@ -17,8 +17,8 @@ RUN pip install mmcv-full==1.3.8 -f https://download.openmmlab.com/mmcv/dist/cu1
 
 # Install MMDetection
 RUN conda clean --all
-RUN git clone https://github.com/open-mmlab/mmdetection.git /mmdetection
-WORKDIR /mmdetection
+RUN git clone https://github.com/shinya7y/UniverseNet.git /UniverseNet
+WORKDIR /UniverseNet
 ENV FORCE_CUDA="1"
 RUN pip install -r requirements/build.txt
 RUN pip install --no-cache-dir -e .
diff --git a/docs/envs/README.md b/docs/envs/README.md
@@ -1,5 +1,14 @@
 ## Testing environment
 
+### UniverseNet 21.09 (Sept. 2021)
+
+- Ubuntu 18.04
+- Python 3.7
+- PyTorch 1.7.1
+- CUDA 10.2
+- GCC 7.5
+- mmcv-full 1.3.9
+
 ### UniverseNet 20.10 (Oct. 2020)
 
 - Ubuntu 18.04

diff --git a/docs/envs/conda_env_universenet_2109.yaml b/docs/envs/conda_env_universenet_2109.yaml
@@ -0,0 +1,103 @@
+name: univ2109
+channels:
+  - pytorch
+  - conda-forge
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=4.5=1_gnu
+  - asynctest=0.13.0=py_0
+  - attrs=21.2.0=pyhd3eb1b0_0
+  - blas=1.0=mkl
+  - ca-certificates=2021.7.5=h06a4308_1
+  - certifi=2021.5.30=py37h06a4308_0
+  - cudatoolkit=10.2.89=hfd86e86_1
+  - cython=0.29.23=py37h2531618_0
+  - flake8=3.9.2=pyhd3eb1b0_0
+  - freetype=2.10.4=h5ab3b9f_0
+  - importlib-metadata=3.10.0=py37h06a4308_0
+  - importlib_metadata=3.10.0=hd3eb1b0_0
+  - iniconfig=1.1.1=pyhd3eb1b0_0
+  - intel-openmp=2021.2.0=h06a4308_610
+  - isort=4.3.21=py37_0
+  - jpeg=9b=h024ee3a_2
+  - lcms2=2.12=h3be6417_0
+  - ld_impl_linux-64=2.35.1=h7274673_9
+  - libffi=3.3=he6710b0_2
+  - libgcc-ng=9.3.0=h5101ec6_17
+  - libgfortran-ng=7.5.0=ha8ba4b0_17
+  - libgfortran4=7.5.0=ha8ba4b0_17
+  - libgomp=9.3.0=h5101ec6_17
+  - libpng=1.6.37=hbc83047_0
+  - libprotobuf=3.15.8=h780b84a_0
+  - libstdcxx-ng=9.3.0=hd4cf53a_17
+  - libtiff=4.2.0=h85742a9_0
+  - libuv=1.40.0=h7b6447c_0
+  - libwebp-base=1.2.0=h27cfd23_0
+  - lz4-c=1.9.3=h2531618_0
+  - mccabe=0.6.1=py37_1
+  - mkl=2021.2.0=h06a4308_296
+  - mkl-service=2.3.0=py37h27cfd23_1
+  - mkl_fft=1.3.0=py37h42c9631_2
+  - mkl_random=1.2.1=py37ha9443f7_2
+  - more-itertools=8.8.0=pyhd3eb1b0_0
+  - ncurses=6.2=he6710b0_1
+  - ninja=1.10.2=hff7bd54_1
+  - numpy=1.20.2=py37h2d18471_0
+  - numpy-base=1.20.2=py37hfae3a4d_0
+  - olefile=0.46=py37_0
+  - onnx=1.9.0=py37h284874a_0
+  - openjpeg=2.3.0=h05c96fa_1
+  - openssl=1.1.1k=h27cfd23_0
+  - packaging=21.0=pyhd3eb1b0_0
+  - pillow=8.3.1=py37h2c7a002_0
+  - pip=21.1.3=py37h06a4308_0
+  - pluggy=0.13.1=py37h06a4308_0
+  - protobuf=3.15.8=py37hcd2ae1e_0
+  - py=1.10.0=pyhd3eb1b0_0
+  - pycodestyle=2.7.0=pyhd3eb1b0_0
+  - pyflakes=2.3.1=pyhd3eb1b0_0
+  - pyparsing=2.4.7=pyhd3eb1b0_0
+  - pytest=6.2.4=py37h06a4308_2
+  - python=3.7.10=h12debd9_4
+  - python_abi=3.7=2_cp37m
+  - pytorch=1.7.1=py3.7_cuda10.2.89_cudnn7.6.5_0
+  - readline=8.1=h27cfd23_0
+  - scipy=1.6.2=py37had2a1c9_1
+  - setuptools=52.0.0=py37h06a4308_0
+  - six=1.16.0=pyhd3eb1b0_0
+  - sqlite=3.36.0=hc218d9a_0
+  - tk=8.6.10=hbc83047_0
+  - toml=0.10.2=pyhd3eb1b0_0
+  - torchvision=0.8.2=py37_cu102
+  - typing-extensions=3.10.0.0=hd3eb1b0_0
+  - typing_extensions=3.10.0.0=pyh06a4308_0
+  - wheel=0.36.2=pyhd3eb1b0_0
+  - xz=5.2.5=h7b6447c_0
+  - zipp=3.5.0=pyhd3eb1b0_0
+  - zlib=1.2.11=h7b6447c_3
+  - zstd=1.4.9=haebb681_0
+  - pip:
+    - addict==2.4.0
+    - appdirs==1.4.4
+    - cfgv==3.3.0
+    - cycler==0.10.0
+    - distlib==0.3.2
+    - filelock==3.0.12
+    - flatbuffers==2.0
+    - identify==2.2.11
+    - kiwisolver==1.3.1
+    - matplotlib==3.4.2
+    - mmcv-full==1.3.9
+    - nodeenv==1.6.0
+    - onnxruntime==1.8.1
+    - opencv-python==4.5.2.54
+    - pre-commit==2.13.0
+    - pycocotools==2.0.2
+    - python-dateutil==2.8.1
+    - pyyaml==5.4.1
+    - terminaltables==3.1.0
+    - timm==0.4.12
+    - virtualenv==20.4.7
+    - xdoctest==0.15.5
+    - yapf==0.31.0
diff --git a/docs/get_started.md b/docs/get_started.md
@@ -117,8 +117,8 @@ Or, you can install MMDetection manually:
 2. Clone the MMDetection repository.
 
     ```shell
-    git clone https://github.com/open-mmlab/mmdetection.git
-    cd mmdetection
+    git clone https://github.com/shinya7y/UniverseNet.git
+    cd UniverseNet
     ```
 
 3. Install build requirements and then install MMDetection.
@@ -207,8 +207,8 @@ conda install pytorch==1.6.0 torchvision==0.7.0 cudatoolkit=10.1 -c pytorch -y
 pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu101/torch1.6.0/index.html
 
 # install mmdetection
-git clone https://github.com/open-mmlab/mmdetection.git
-cd mmdetection
+git clone https://github.com/shinya7y/UniverseNet.git
+cd UniverseNet
 pip install -r requirements/build.txt
 pip install -v -e .
 ```