Bump version to v0.8.0 (#191)

* Bump version to v0.8.0 * Unify mmcv min and max version * Revise the min version of mmcv to 1.3.17
open-mmlab · Jun 1, 2022 · d0d0c98 · d0d0c98
1 parent ff99666
commit d0d0c98
Show file tree

Hide file tree

Showing 6 changed files with 25 additions and 29 deletions.
diff --git a/README.md b/README.md
@@ -42,6 +42,10 @@ https://user-images.githubusercontent.com/62529255/144362861-e794b404-c48f-4ebe-
   A suite of differentiale visualization tools for human parametric model rendering (including part segmentation, depth map and point clouds) and conventional 2D/3D keypoints are available.
 
 ## News
+- 2022-05-31: MMHuman3D [v0.8.0](https://github.com/open-mmlab/mmhuman3d/releases/tag/v0.8.0) is released. Major updates include:
+  - Support SmoothNet (added by paper authors)
+  - Fix circular import and up to 2.5x speed up in module initialization
+  - Add documentations in Chinese
 - 2022-04-30: MMHuman3D [v0.7.0](https://github.com/open-mmlab/mmhuman3d/releases/tag/v0.7.0) is released. Major updates include:
   - Support PARE (better than the official implementation)
   - Support DeciWatch (added by paper authors)
@@ -51,12 +55,6 @@ https://user-images.githubusercontent.com/62529255/144362861-e794b404-c48f-4ebe-
   - Add HumanDataCache that requires 96% less RAM during training
   - Refactor differentiable renderers and support UV map rendering
   - Support slice/concat operations for HumanData
-- 2022-02-11: MMHuman3D [v0.5.0](https://github.com/open-mmlab/mmhuman3d/releases/tag/v0.5.0) is released. Major updates include:
-  - Support new data structure SMC for new dataset HuMMan, which will soon be released
-  - Support for multi-GPU training/testing without slurm
-  - Support training-time validation and additional metrics such as PVE
-  - Bug fixes in data augmentation for more stable training
-  - Stronger HybrIK baseline (PA-MPJPE 49.02 on 3DPW)
 
 ## Benchmark and Model Zoo
 

diff --git a/README_CN.md b/README_CN.md
@@ -42,6 +42,10 @@ https://user-images.githubusercontent.com/62529255/144362861-e794b404-c48f-4ebe-
   一整套可微的可视化工具支持人体参数化模型的渲染（包括部分分割，深度图以及点云）和传统 2D/3D 关键点的可视化。
 
 ## 最新进展
+- 2022-05-31: MMHuman3D [v0.8.0](https://github.com/open-mmlab/mmhuman3d/releases/tag/v0.8.0) 已经发布. 主要更新包括:
+  - 支持 SmoothNet（由论文作者添加）
+  - 修复循环引用问题，获得最多2.5倍速度提升
+  - 增加中文版文档
 - 2022-04-30: MMHuman3D [v0.7.0](https://github.com/open-mmlab/mmhuman3d/releases/tag/v0.7.0) 已经发布. 主要更新包括:
   - 支持PARE算法 (优于官方实现）
   - 支持DeciWatch（由论文作者添加）
@@ -51,12 +55,6 @@ https://user-images.githubusercontent.com/62529255/144362861-e794b404-c48f-4ebe-
   - 增加HumanDataCache，大幅削减(96%)训练时内存占用
   - 重构可微渲染器并支持UV map渲染
   - HumanData支持slice/concat操作
-- 2022-02-11: MMHuman3D [v0.5.0](https://github.com/open-mmlab/mmhuman3d/releases/tag/v0.5.0) 已经发布. 主要更新包括:
-  - 支持新的数据结构SMC以及即将发布的新数据集HuMMan
-  - 支持在非slurm环境下的多GPU训练/测试
-  - 支持训练时验证及更多测试指标如PVE
-  - 修复数据增强的问题使训练稳定
-  - 更强的HybrIK基线（3DPW上达到PA-MPJPE 49.02）
 
 ## 基准与模型库
 

diff --git a/docs/install.md b/docs/install.md
@@ -18,12 +18,12 @@
 - CUDA 9.2+
 - GCC 5+
 - PyTorch3D 0.4+
-- [MMCV](https://github.com/open-mmlab/mmcv) (Please install mmcv-full>=1.3.13, <=1.5.0 for GPU)
+- [MMCV](https://github.com/open-mmlab/mmcv) (Please install mmcv-full>=1.3.17,<1.6.0 for GPU)
 
 Optional:
 - [MMPOSE](https://github.com/open-mmlab/mmpose) (Only for demo.)
 - [MMDETECTION](https://github.com/open-mmlab/mmdetection) (Only for demo.)
-- [MMTRACKING](https://github.com/open-mmlab/mmtracking) (Only for multi-person demo. If you use mmtrack, please install mmcls<1.18.0, mmcv-full>=1.3.16, <1.4.0 for GPU)
+- [MMTRACKING](https://github.com/open-mmlab/mmtracking) (Only for multi-person demo. If you use mmtrack, please install mmcls<1.18.0, mmcv-full>=1.3.16,<1.6.0 for GPU
 
 ## Prepare environment
 
@@ -95,13 +95,13 @@ Please replace `{torch_version}` in the url to your desired one.
 
 For GPU:
  ```shell
- pip install "mmcv-full>=1.3.13,<1.4.0" -f https://download.openmmlab.com/mmcv/dist/{cu_version}/{torch_version}/index.html
+ pip install "mmcv-full>=1.3.17,<1.6.0" -f https://download.openmmlab.com/mmcv/dist/{cu_version}/{torch_version}/index.html
  ```
 Please replace `{cu_version}` and `{torch_version}` in the url to your desired one.
 
 For example, to install mmcv-full with CUDA 10.2 and PyTorch 1.8.0, use the following command:
 ```shell
-pip install "mmcv-full>=1.3.13,<1.4.0" -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
+pip install "mmcv-full>=1.3.17,<1.6.0" -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
 ```
 
 See [here](https://mmcv.readthedocs.io/en/latest/get_started/installation.html) for different versions of MMCV compatible to different PyTorch and CUDA versions.
@@ -110,7 +110,7 @@ For more version download link, refer to [openmmlab-download](https://download.o
 Optionally you can choose to compile mmcv from source by the following command
 
 ```shell
-git clone https://github.com/open-mmlab/mmcv.git -b v1.3.13
+git clone https://github.com/open-mmlab/mmcv.git -b v1.3.17
 cd mmcv
 MMCV_WITH_OPS=1 pip install -e .  # package mmcv-full, which contains cuda ops, will be installed after this step
 # OR pip install -e .  # package mmcv, which contains no cuda ops, will be installed after this step
@@ -192,7 +192,7 @@ conda install pytorch3d -c pytorch3d
 
 # install mmcv-full
 
-pip install "mmcv-full>=1.3.13,<1.4.0" -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
+pip install "mmcv-full>=1.3.17,<1.6.0" -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
 
 
 # Optional

diff --git a/docs_zh-CN/install.md b/docs_zh-CN/install.md
@@ -14,12 +14,12 @@
 - CUDA 9.2+
 - GCC 5+
 - PyTorch3D 0.4+
-- [MMCV](https://github.com/open-mmlab/mmcv) (请安装用于GPU的mmcv-full, 要求版本>=1.3.13, <=1.5.0)
+- [MMCV](https://github.com/open-mmlab/mmcv) (请安装用于GPU的mmcv-full, 要求版本>=1.3.17,<1.6.0)
 
 Optional:
 - [MMPose](https://github.com/open-mmlab/mmpose) (只用于演示)
 - [MMDetection](https://github.com/open-mmlab/mmdetection) (只用于演示)
-- [MMTracking](https://github.com/open-mmlab/mmtracking) (只用于多人的演示。请安装 mmcls<1.18.0, mmcv-full>=1.3.16, <1.4.0)
+- [MMTracking](https://github.com/open-mmlab/mmtracking) (只用于多人的演示。请安装 mmcls<1.18.0, mmcv-full>=1.3.16,<1.6.0)
 
 ## 准备环境
 
@@ -91,21 +91,21 @@ pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/cpu/{torch_ver
 
 对于 GPU:
  ```shell
- pip install "mmcv-full>=1.3.13,<1.4.0" -f https://download.openmmlab.com/mmcv/dist/{cu_version}/{torch_version}/index.html
+ pip install "mmcv-full>=1.3.17,<1.6.0" -f https://download.openmmlab.com/mmcv/dist/{cu_version}/{torch_version}/index.html
  ```
 使用您的cuda版本号和pytorch版本号替换`{cu_version}`和`{torch_version}`
 
 以在CUDA 10.2和PyTorch 1.8.0的环境下, 安装mmcv-full为例:
 ```shell
-pip install "mmcv-full>=1.3.13,<1.4.0" -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
+pip install "mmcv-full>=1.3.17,<1.6.0" -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
 ```
 您可以从[这里](https://mmcv.readthedocs.io/en/latest/get_started/installation.html)查看适配于不同CUDA版本和PyTorch版本的MMCV.
 更多版本的下载信息，请参考[openmmlab-download](https://download.openmmlab.com/mmcv/dist/index.html).
 
 您也可以使用如下命令，从源码编译mmcv
 
 ```shell
-git clone https://github.com/open-mmlab/mmcv.git -b v1.3.13
+git clone https://github.com/open-mmlab/mmcv.git -b v1.3.17
 cd mmcv
 MMCV_WITH_OPS=1 pip install -e .  # 执行此步骤会安装包含cuda算子的mmcv-full
 # 或者运行 pip install -e .  # 执行此步骤会安装不包含cuda算子的mmcv
@@ -187,7 +187,7 @@ conda install pytorch3d -c pytorch3d
 
 # 安装 mmcv-full
 
-pip install "mmcv-full>=1.3.13,<1.4.0" -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
+pip install "mmcv-full>=1.3.17,<1.6.0" -f https://download.openmmlab.com/mmcv/dist/cu102/torch1.8.0/index.html
 
 
 # 可选

diff --git a/mmhuman3d/__init__.py b/mmhuman3d/__init__.py
@@ -15,14 +15,14 @@ def digit_version(version_str):
     return digit_version
 
 
-mmcv_minimum_version = '1.3.1'
-mmcv_maximum_version = '1.5.1'
+mmcv_minimum_version = '1.3.17'
+mmcv_maximum_version = '1.6.0'
 mmcv_version = digit_version(mmcv.__version__)
 
 
 assert (mmcv_version >= digit_version(mmcv_minimum_version)
-        and mmcv_version <= digit_version(mmcv_maximum_version)), \
+        and mmcv_version < digit_version(mmcv_maximum_version)), \
     f'MMCV=={mmcv.__version__} is used but incompatible. ' \
-    f'Please install mmcv>={mmcv_minimum_version}, <={mmcv_maximum_version}.'
+    f'Please install mmcv>={mmcv_minimum_version}, <{mmcv_maximum_version}.'
 
 __all__ = ['__version__']
diff --git a/mmhuman3d/version.py b/mmhuman3d/version.py
@@ -1,6 +1,6 @@
 # Copyright (c) Open-MMLab. All rights reserved.
 
-__version__ = '0.7.0'
+__version__ = '0.8.0'
 
 
 def parse_version_info(version_str):