Skip to content

Commit 2028b0c

Browse files
hhaAndroidhhaAndroid
authored
Bump version to v2.14.0 (#5466)
* Bump version to v2.14.0 * update * update * update * update version
1 parent 3438a50 commit 2028b0c

File tree

7 files changed

+51
-7
lines changed

7 files changed

+51
-7
lines changed

‎README.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -44,7 +44,7 @@ This project is released under the [Apache 2.0 license](LICENSE).
4444

4545
## Changelog
4646

47-
v2.13.0 was released in 01/06/2021.
47+
v2.14.0 was released in 29/06/2021.
4848
Please refer to [changelog.md](docs/changelog.md) for details and release history.
4949
A comparison between v1.x and v2.0 codebases can be found in [compatibility.md](docs/compatibility.md).
5050

‎README_zh-CN.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -44,7 +44,7 @@ v1.x 的历史版本支持 PyTorch 1.1 到 1.4，但是我们强烈建议用户
4444

4545
## 更新日志
4646

47-
最新的月度版本 v2.13.0 在 2021.06.01 发布。
47+
最新的月度版本 v2.14.0 在 2021.06.29 发布。
4848
如果想了解更多版本更新细节和历史信息，请阅读[更新日志](docs/changelog.md)。
4949
在[兼容性说明文档](docs/compatibility.md)中我们提供了 1.x 和 2.0 版本的详细比较。
5050

‎docker/serve/Dockerfile

+2-2
Original file line numberDiff line numberDiff line change
@@ -3,8 +3,8 @@ ARG CUDA="10.1"
33
ARG CUDNN="7"
44
FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel
55

6-
ARG MMCV="1.2.7"
7-
ARG MMDET="2.13.0"
6+
ARG MMCV="1.3.8"
7+
ARG MMDET="2.14.0"
88

99
ENV PYTHONUNBUFFERED TRUE
1010

‎docs/changelog.md

+43
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,48 @@
11
## Changelog
22

3+
# v2.14.0 (29/6/2021)
4+
5+
## Highlights
6+
7+
- Add `simple_test` to dense heads to improve the consistency of single-stage and two-stage detectors
8+
- Revert the `test_mixins` to single image test to improve efficiency and readability
9+
- Add Faster R-CNN and Mask R-CNN config using multi-scale training with 3x schedule
10+
11+
## New Features
12+
13+
- Support pretrained models from MoCo v2 and SwAV (#5286)
14+
- Add Faster R-CNN and Mask R-CNN config using multi-scale training with 3x schedule (#5179, #5233)
15+
- Add `reduction_override` in MSELoss (#5437)
16+
- Stable support of exporting DETR to ONNX with dynamic shapes and batch inference (#5168)
17+
- Stable support of exporting PointRend to ONNX with dynamic shapes and batch inference (#5440)
18+
19+
## Bug Fixes
20+
21+
- Fix size mismatch bug in `multiclass_nms` (#4980)
22+
- Fix the import path of `MultiScaleDeformableAttention` (#5338)
23+
- Fix errors in config of GCNet ResNext101 models (#5360)
24+
- Fix Grid-RCNN error when there is no bbox result (#5357)
25+
- Fix errors in `onnx_export` of bbox_head when setting reg_class_agnostic (#5468)
26+
- Fix type error of AutoAssign in the document (#5478)
27+
- Fix web links ending with `.md` (#5315)
28+
29+
## Improvements
30+
31+
- Add `simple_test` to dense heads to improve the consistency of single-stage and two-stage detectors (#5264)
32+
- Add support for mask diagonal flip in TTA (#5403)
33+
- Revert the `test_mixins` to single image test to improve efficiency and readability (#5249)
34+
- Make YOLOv3 Neck more flexible (#5218)
35+
- Refactor SSD to make it more general (#5291)
36+
- Refactor `anchor_generator` and `point_generator` (#5349)
37+
- Allow to configure out the `mask_head` of the HTC algorithm (#5389)
38+
- Delete deprecated warning in FPN (#5311)
39+
- Move `model.pretrained` to `model.backbone.init_cfg` (#5370)
40+
- Make deployment tools more friendly to use (#5280)
41+
- Clarify installation documentation (#5316)
42+
- Add ImageNet Pretrained Models docs (#5268)
43+
- Add FAQ about training loss=nan solution and COCO AP or AR =-1 (# 5312, #5313)
44+
- Change all weight links of http to https (#5328)
45+
346
### v2.13.0 (01/6/2021)
447

548
#### Highlights

‎docs/get_started.md

+2-1
Original file line numberDiff line numberDiff line change
@@ -11,7 +11,8 @@ The compatible MMDetection and MMCV versions are as below. Please install the co
1111

1212
| MMDetection version | MMCV version |
1313
|:-------------------:|:-------------------:|
14-
| master | mmcv-full>=1.3.3, <1.4.0 |
14+
| master | mmcv-full>=1.3.8, <1.4.0 |
15+
| 2.14.0 | mmcv-full>=1.3.8, <1.4.0 |
1516
| 2.13.0 | mmcv-full>=1.3.3, <1.4.0 |
1617
| 2.12.0 | mmcv-full>=1.3.3, <1.4.0 |
1718
| 2.11.0 | mmcv-full>=1.2.4, <1.4.0 |

‎mmdet/__init__.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -15,7 +15,7 @@ def digit_version(version_str):
1515
return digit_version
1616

1717

18-
mmcv_minimum_version = '1.3.2'
18+
mmcv_minimum_version = '1.3.8'
1919
mmcv_maximum_version = '1.4.0'
2020
mmcv_version = digit_version(mmcv.__version__)
2121

‎mmdet/version.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
# Copyright (c) Open-MMLab. All rights reserved.
22

3-
__version__ = '2.13.0'
3+
__version__ = '2.14.0'
44
short_version = __version__
55

66

0 commit comments

Comments
 (0)