[Docs] update dependency version req, dockerfile and change logs for 0.2.1 (#331)

* update pytorch req and dockerfile * Update dependency requirement * update readme for 0.2.1 * update change log * update release date
2021-07-20 23:18:47 +08:00 · 2021-07-20 23:18:47 +08:00 · d8cb48ab46
parent 4f7270e574
commit d8cb48ab46
6 changed files with 91 additions and 41 deletions
--- a/README.md
+++ b/README.md
@ -16,7 +16,7 @@ English | [简体中文](README_zh-CN.md)

 MMOCR is an open-source toolbox based on PyTorch and mmdetection for text detection, text recognition, and the corresponding downstream tasks including key information extraction. It is part of the [OpenMMLab](https://openmmlab.com/) project.

-The main branch works with **PyTorch 1.5+**.
+The main branch works with **PyTorch 1.6+**.

 Documentation: https://mmocr.readthedocs.io/en/latest/.

@ -105,7 +105,7 @@ If you find this project useful in your research, please consider cite:

 ## Changelog

-v0.2.0 was released in 2021-5-18.
+v0.2.1 was released in 2021-7-20.


 ## Installation
--- a/README_zh-CN.md
+++ b/README_zh-CN.md
@ -16,7 +16,7 @@

 MMOCR 是基于 PyTorch 和 mmdetection 的开源工具箱，专注于文本检测，文本识别以及相应的下游任务，如关键信息提取。 它是 OpenMMLab 项目的一部分。

-主分支目前支持 **PyTorch 1.5 以上**的版本。
+主分支目前支持 **PyTorch 1.6 以上**的版本。

 文档：https://mmocr.readthedocs.io/en/latest/。

@ -105,7 +105,7 @@ MMOCR 是基于 PyTorch 和 mmdetection 的开源工具箱，专注于文本检

 ## 更新日志

-最新的月度版本 v0.2.0 在 2021.05.18 发布。
+最新的月度版本 v0.2.1 在 2021.07.20 发布。


 ## 安装
--- a/docker/Dockerfile
+++ b/docker/Dockerfile
@ -1,4 +1,4 @@
-ARG PYTORCH="1.5"
+ARG PYTORCH="1.6.0"
 ARG CUDA="10.1"
 ARG CUDNN="7"

@ -13,9 +13,9 @@ RUN apt-get update && apt-get install -y git ninja-build libglib2.0-0 libsm6 lib
    && rm -rf /var/lib/apt/lists/*

 RUN conda clean --all
-RUN pip install mmcv-full==1.3.4+torch1.5.0+cu101 -f https://download.openmmlab.com/mmcv/dist/index.html
+RUN pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu101/torch1.6.0/index.html

-RUN pip install mmdet==2.11.0
+RUN pip install mmdet==2.13.0

 RUN git clone https://github.com/open-mmlab/mmocr.git /mmocr
 WORKDIR /mmocr
--- a/docs/changelog.md
+++ b/docs/changelog.md
@ -1,5 +1,55 @@
 # Changelog

+## v0.2.1 (20/7/2021)
+
+**Highlights**
+1. Upgrade to use MMCV-full **>= 1.3.8** and MMDetection **>= 2.13.0** for latest features
+2. Add ONNX and TensorRT export tool, supporting the deployment of DBNet, PSENet, PANet and CRNN (experimental) [#278](https://github.com/open-mmlab/mmocr/pull/278), [#291](https://github.com/open-mmlab/mmocr/pull/291), [#300](https://github.com/open-mmlab/mmocr/pull/300), [#328](https://github.com/open-mmlab/mmocr/pull/328)
+3. Unified parameter initialization method which uses init_cfg in config files [#365](https://github.com/open-mmlab/mmocr/pull/365)
+
+**New Features**
+
+- Support TextOCR dataset [#293](https://github.com/open-mmlab/mmocr/pull/293)
+- Support Total-Text dataset [#266](https://github.com/open-mmlab/mmocr/pull/266), [#273](https://github.com/open-mmlab/mmocr/pull/273), [#357](https://github.com/open-mmlab/mmocr/pull/357)
+- Support grouping text detection box into lines [#290](https://github.com/open-mmlab/mmocr/pull/290), [#304](https://github.com/open-mmlab/mmocr/pull/304)
+- Add benchmark_processing script that benchmarks data loading process [#261](https://github.com/open-mmlab/mmocr/pull/261)
+- Add SynthText preprocessor for text recognition models [#351](https://github.com/open-mmlab/mmocr/pull/351), [#361](https://github.com/open-mmlab/mmocr/pull/361)
+- Support batch inference during testing [#310](https://github.com/open-mmlab/mmocr/pull/310)
+- Add user-friendly OCR inference script [#366](https://github.com/open-mmlab/mmocr/pull/366)
+
+**Bug Fixes**
+
+- Fix improper class ignorance in SDMGR Loss [#221](https://github.com/open-mmlab/mmocr/pull/221)
+- Fix potential numerical zero division error in DRRG [#224](https://github.com/open-mmlab/mmocr/pull/224)
+- Fix installing requirements with pip and mim [#242](https://github.com/open-mmlab/mmocr/pull/242)
+- Fix dynamic input error of DBNet [#269](https://github.com/open-mmlab/mmocr/pull/269)
+- Fix space parsing error in LineStrParser [#285](https://github.com/open-mmlab/mmocr/pull/285)
+- Fix textsnake decode error [#264](https://github.com/open-mmlab/mmocr/pull/264)
+- Correct isort setup [#288](https://github.com/open-mmlab/mmocr/pull/288)
+- Fix a bug in SDMGR config [#316](https://github.com/open-mmlab/mmocr/pull/316)
+- Fix kie_test_img for KIE nonvisual [#319](https://github.com/open-mmlab/mmocr/pull/319)
+- Fix metafiles [#342](https://github.com/open-mmlab/mmocr/pull/342)
+- Fix different device problem in FCENet [#334](https://github.com/open-mmlab/mmocr/pull/334)
+- Ignore improper tailing empty characters in annotation files [#358](https://github.com/open-mmlab/mmocr/pull/358)
+- Docs fixes [#247](https://github.com/open-mmlab/mmocr/pull/247), [#255](https://github.com/open-mmlab/mmocr/pull/255), [#265](https://github.com/open-mmlab/mmocr/pull/265), [#267](https://github.com/open-mmlab/mmocr/pull/267), [#268](https://github.com/open-mmlab/mmocr/pull/268), [#270](https://github.com/open-mmlab/mmocr/pull/270), [#276](https://github.com/open-mmlab/mmocr/pull/276), [#287](https://github.com/open-mmlab/mmocr/pull/287), [#330](https://github.com/open-mmlab/mmocr/pull/330), [#355](https://github.com/open-mmlab/mmocr/pull/355), [#367](https://github.com/open-mmlab/mmocr/pull/367)
+- Fix NRTR config [#356](https://github.com/open-mmlab/mmocr/pull/356), [#370](https://github.com/open-mmlab/mmocr/pull/370)
+
+**Improvements**
+
+- Add backend for resizeocr [#244](https://github.com/open-mmlab/mmocr/pull/244)
+- Skip image processing pipelines in SDMGR novisual [#260](https://github.com/open-mmlab/mmocr/pull/260)
+- Speedup DBNet [#263](https://github.com/open-mmlab/mmocr/pull/263)
+- Update mmcv installation method in workflow [#323](https://github.com/open-mmlab/mmocr/pull/323)
+- Add part of Chinese documentations [#353](https://github.com/open-mmlab/mmocr/pull/353), [#362](https://github.com/open-mmlab/mmocr/pull/362)
+- Add support for ConcatDataset with two workflows [#348](https://github.com/open-mmlab/mmocr/pull/348)
+- Add list_from_file and list_to_file utils [#226](https://github.com/open-mmlab/mmocr/pull/226)
+- Speed up sort_vertex [#239](https://github.com/open-mmlab/mmocr/pull/239)
+- Support distributed evaluation of KIE [#234](https://github.com/open-mmlab/mmocr/pull/234)
+- Add pretrained FCENet on IC15 [#258](https://github.com/open-mmlab/mmocr/pull/258)
+- Support CPU for OCR demo [#227](https://github.com/open-mmlab/mmocr/pull/227)
+- Avoid extra image pre-processing steps [#375](https://github.com/open-mmlab/mmocr/pull/375)
+
+
 ## v0.2.0 (18/5/2021)

 **Highlights**
--- a/docs/install.md
+++ b/docs/install.md
@ -4,25 +4,25 @@

 - Linux (Windows is not officially supported)
 - Python 3.7
- PyTorch 1.5 or higher
- torchvision 0.6.0
+- PyTorch 1.6 or higher
+- torchvision 0.7.0
 - CUDA 10.1
 - NCCL 2
 - GCC 5.4.0 or higher
- [MMCV](https://mmcv.readthedocs.io/en/latest/#installation) 1.3.4
- [MMDetection](https://mmdetection.readthedocs.io/en/latest/#installation) 2.11.0
+- [MMCV](https://mmcv.readthedocs.io/en/latest/#installation) >= 1.3.8
+- [MMDetection](https://mmdetection.readthedocs.io/en/latest/#installation) >= 2.13.0

 We have tested the following versions of OS and softwares:

 - OS: Ubuntu 16.04
 - CUDA: 10.1
 - GCC(G++): 5.4.0
- MMCV 1.3.4
- MMDetection 2.11.0
- PyTorch 1.5
- torchvision 0.6.0
+- MMCV 1.3.8
+- MMDetection 2.13.0
+- PyTorch 1.6.0
+- torchvision 0.7.0

-MMOCR depends on Pytorch and mmdetection.
+MMOCR depends on PyTorch and mmdetection.

 ## Step-by-Step Installation Instructions

@ -36,7 +36,7 @@ conda activate open-mmlab
 b. Install PyTorch and torchvision following the [official instructions](https://pytorch.org/), e.g.,

 ```shell
-conda install pytorch==1.5.0 torchvision==0.6.0 cudatoolkit=10.1 -c pytorch
+conda install pytorch==1.6.0 torchvision==0.7.0 cudatoolkit=10.1 -c pytorch
 ```
 Note: Make sure that your compilation CUDA version and runtime CUDA version match.
 You can check the supported CUDA version for precompiled packages on the [PyTorch website](https://pytorch.org/).
@ -53,9 +53,9 @@ Please replace ``{cu_version}`` and ``{torch_version}`` in the url to your desir
 ```shell
 pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu110/torch1.7.0/index.html
 ```
-Note that mmocr 0.2.0 or later require mmcv 1.3.4 or later.
+Note that mmocr 0.2.1 or later requires mmcv 1.3.8 or later.

-If it compiles during installation, then please check that the cuda version and pytorch version **exactly"" matches the version in the mmcv-full installation command. For example, pytorch 1.7.0 and 1.7.1 are treated differently.
+If it compiles during installation, then please check that the cuda version and pytorch version **exactly** matches the version in the mmcv-full installation command. For example, pytorch 1.7.0 and 1.7.1 are treated differently.

 See official [installation](https://github.com/open-mmlab/mmcv#installation) for different versions of MMCV compatible to different PyTorch and CUDA versions.

@ -65,7 +65,7 @@ d. Install [mmdet](https://github.com/open-mmlab/mmdetection.git), we recommend
 See [here](https://pypi.org/project/mmdet/) for different versions of `mmdet`.

 ```shell
-pip install mmdet==2.11.0
+pip install mmdet
 ```

 Optionally you can choose to install `mmdet` following the official [installation](https://github.com/open-mmlab/mmdetection/blob/master/docs/get_started.md).
@ -95,13 +95,13 @@ conda create -n open-mmlab python=3.7 -y
 conda activate open-mmlab

 # install latest pytorch prebuilt with the default prebuilt CUDA version (usually the latest)
-conda install pytorch==1.5.0 torchvision==0.6.0 cudatoolkit=10.1 -c pytorch
+conda install pytorch==1.6.0 torchvision==0.7.0 cudatoolkit=10.1 -c pytorch

 # install the latest mmcv-full
-pip install mmcv-full==1.3.4
+pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu101/torch1.6.0/index.html

 # install mmdetection
-pip install mmdet==2.11.0
+pip install mmdet

 # install mmocr
 git clone https://github.com/open-mmlab/mmocr.git
@ -117,7 +117,7 @@ export PYTHONPATH=$(pwd):$PYTHONPATH
 We provide a [Dockerfile](https://github.com/open-mmlab/mmocr/blob/master/docker/Dockerfile) to build an image.

 ```shell
-# build an image with PyTorch 1.5, CUDA 10.1
+# build an image with PyTorch 1.6, CUDA 10.1
 docker build -t mmocr docker/
 ```

--- a/docs_zh_CN/install.md
+++ b/docs_zh_CN/install.md
@ -4,25 +4,25 @@

 - Linux (Windows is not officially supported)
 - Python 3.7
- PyTorch 1.5 or higher
- torchvision 0.6.0
+- PyTorch 1.6 or higher
+- torchvision 0.7.0
 - CUDA 10.1
 - NCCL 2
 - GCC 5.4.0 or higher
- [MMCV](https://mmcv.readthedocs.io/en/latest/#installation) 1.3.4
- [MMDetection](https://mmdetection.readthedocs.io/en/latest/#installation) 2.11.0
+- [MMCV](https://mmcv.readthedocs.io/en/latest/#installation) >= 1.3.8
+- [MMDetection](https://mmdetection.readthedocs.io/en/latest/#installation) >= 2.13.0

 We have tested the following versions of OS and softwares:

 - OS: Ubuntu 16.04
 - CUDA: 10.1
 - GCC(G++): 5.4.0
- MMCV 1.3.4
- MMDetection 2.11.0
- PyTorch 1.5
- torchvision 0.6.0
+- MMCV 1.3.8
+- MMDetection 2.13.0
+- PyTorch 1.6.0
+- torchvision 0.7.0

-MMOCR depends on Pytorch and mmdetection.
+MMOCR depends on PyTorch and mmdetection.

 ## Step-by-Step Installation Instructions

@ -36,7 +36,7 @@ conda activate open-mmlab
 b. Install PyTorch and torchvision following the [official instructions](https://pytorch.org/), e.g.,

 ```shell
-conda install pytorch==1.5.0 torchvision==0.6.0 cudatoolkit=10.1 -c pytorch
+conda install pytorch==1.6.0 torchvision==0.7.0 cudatoolkit=10.1 -c pytorch
 ```
 Note: Make sure that your compilation CUDA version and runtime CUDA version match.
 You can check the supported CUDA version for precompiled packages on the [PyTorch website](https://pytorch.org/).
@ -53,9 +53,9 @@ Please replace ``{cu_version}`` and ``{torch_version}`` in the url to your desir
 ```shell
 pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu110/torch1.7.0/index.html
 ```
-Note that mmocr 0.2.0 or later require mmcv 1.3.4 or later.
+Note that mmocr 0.2.1 or later requires mmcv 1.3.8 or later.

-If it compiles during installation, then please check that the cuda version and pytorch version **exactly"" matches the version in the mmcv-full installation command. For example, pytorch 1.7.0 and 1.7.1 are treated differently.
+If it compiles during installation, then please check that the cuda version and pytorch version **exactly** matches the version in the mmcv-full installation command. For example, pytorch 1.7.0 and 1.7.1 are treated differently.

 See official [installation](https://github.com/open-mmlab/mmcv#installation) for different versions of MMCV compatible to different PyTorch and CUDA versions.

@ -65,7 +65,7 @@ d. Install [mmdet](https://github.com/open-mmlab/mmdetection.git), we recommend
 See [here](https://pypi.org/project/mmdet/) for different versions of `mmdet`.

 ```shell
-pip install mmdet==2.11.0
+pip install mmdet
 ```

 Optionally you can choose to install `mmdet` following the official [installation](https://github.com/open-mmlab/mmdetection/blob/master/docs/get_started.md).
@ -95,13 +95,13 @@ conda create -n open-mmlab python=3.7 -y
 conda activate open-mmlab

 # install latest pytorch prebuilt with the default prebuilt CUDA version (usually the latest)
-conda install pytorch==1.5.0 torchvision==0.6.0 cudatoolkit=10.1 -c pytorch
+conda install pytorch==1.6.0 torchvision==0.7.0 cudatoolkit=10.1 -c pytorch

 # install the latest mmcv-full
-pip install mmcv-full==1.3.4
+pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cu101/torch1.6.0/index.html

 # install mmdetection
-pip install mmdet==2.11.0
+pip install mmdet

 # install mmocr
 git clone https://github.com/open-mmlab/mmocr.git
@ -117,7 +117,7 @@ export PYTHONPATH=$(pwd):$PYTHONPATH
 We provide a [Dockerfile](https://github.com/open-mmlab/mmocr/blob/master/docker/Dockerfile) to build an image.

 ```shell
-# build an image with PyTorch 1.5, CUDA 10.1
+# build an image with PyTorch 1.6, CUDA 10.1
 docker build -t mmocr docker/
 ```