Skip to content

Commit 27487fd

Browse files
hhaAndroidhhaAndroid
authored
Merge pull request #331 from open-mmlab/dev
Bump version to v0.2.0
2 parents 178b0bf + 4e8dac5 commit 27487fd

File tree

133 files changed

+11205
-886
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

133 files changed

+11205
-886
lines changed

‎README.md

+9-13
Original file line numberDiff line numberDiff line change
@@ -71,19 +71,14 @@ And the figure of P6 model is in [model_design.md](docs/en/algorithm_description
7171

7272
## What's New
7373

74-
💎 **v0.1.3** was released on 10/11/2022:
74+
💎 **v0.2.0** was released on 1/12/2022:
7575

76-
1. Fix training failure when saving best weights based on mmengine 0.3.1
77-
2. Fix `add_dump_metric` error based on mmdet 3.0.0rc3
78-
79-
💎 **v0.1.2** was released on 3/11/2022:
80-
81-
1. Support [YOLOv5/YOLOv6/YOLOX/RTMDet deployments](https://github.com/open-mmlab/mmyolo/blob/main/configs/deploy) for ONNXRuntime and TensorRT
82-
2. Support [YOLOv6](https://github.com/open-mmlab/mmyolo/blob/main/configs/yolov6) s/t/n model training
83-
3. YOLOv5 supports [P6 model training which can input 1280-scale images](https://github.com/open-mmlab/mmyolo/blob/main/configs/yolov5)
84-
4. YOLOv5 supports [VOC dataset training](https://github.com/open-mmlab/mmyolo/blob/main/configs/yolov5/voc)
85-
5. Support [PPYOLOE](https://github.com/open-mmlab/mmyolo/blob/main/configs/ppyoloe) and [YOLOv7](https://github.com/open-mmlab/mmyolo/blob/main/configs/yolov7) model inference and official weight conversion
86-
6. Add YOLOv5 replacement [backbone tutorial](https://github.com/open-mmlab/mmyolo/blob/dev/docs/en/advanced_guides/how_to.md#use-backbone-network-implemented-in-other-openmmlab-repositories) in How-to documentation
76+
1. Support [YOLOv7](https://github.com/open-mmlab/mmyolo/tree/dev/configs/yolov7) P5 and P6 model
77+
2. Support [YOLOv6](https://github.com/open-mmlab/mmyolo/blob/dev/configs/yolov6/README.md) ML model
78+
3. Support [Grad-Based CAM and Grad-Free CAM](https://github.com/open-mmlab/mmyolo/blob/dev/demo/boxam_vis_demo.py)
79+
4. Support [large image inference](https://github.com/open-mmlab/mmyolo/blob/dev/demo/large_image_demo.py) based on sahi
80+
5. Add [easydeploy](https://github.com/open-mmlab/mmyolo/blob/dev/projects/easydeploy/README.md) project under the projects folder
81+
6. Add [custom dataset guide](https://github.com/open-mmlab/mmyolo/blob/dev/docs/zh_cn/user_guides/custom_dataset.md)
8782

8883
For release history and update details, please refer to [changelog](https://mmyolo.readthedocs.io/en/latest/notes/changelog.html).
8984

@@ -123,6 +118,7 @@ For different parts from MMDetection, we have also prepared user guides and adva
123118
- [Useful Tools](https://mmdetection.readthedocs.io/en/latest/user_guides/index.html#useful-tools)
124119
- [Visualization](docs/en/user_guides/visualization.md)
125120
- [Useful Tools](docs/en/user_guides/useful_tools.md)
121+
- [Custom Dataset](docs/en/user_guides/custom_dataset.md)
126122

127123
- Algorithm description
128124

@@ -155,8 +151,8 @@ Results and models are available in the [model zoo](docs/en/model_zoo.md).
155151
- [x] [YOLOX](configs/yolox)
156152
- [x] [RTMDet](configs/rtmdet)
157153
- [x] [YOLOv6](configs/yolov6)
154+
- [x] [YOLOv7](configs/yolov7)
158155
- [ ] [PPYOLOE](configs/ppyoloe)(Inference only)
159-
- [ ] [YOLOv7](configs/yolov7)(Inference only)
160156

161157
</details>
162158

‎README_zh-CN.md

+13-14
Original file line numberDiff line numberDiff line change
@@ -71,19 +71,14 @@ P6 模型图详见 [model_design.md](docs/zh_CN/algorithm_descriptions/model_des
7171

7272
## 最新进展
7373

74-
💎 **v0.1.3** 版本已经在 2022.11.10 发布：
74+
💎 **v0.2.0** 版本已经在 2022.12.1 发布：
7575

76-
1. 基于 mmengine 0.3.1 修复保存最好权重时训练失败问题
77-
2. 基于 mmdet 3.0.0rc3 修复 `add_dump_metric` 报错 (#253)
78-
79-
💎 **v0.1.2** 版本已经在 2022.11.3 发布：
80-
81-
1. 支持 ONNXRuntime 和 TensorRT 的 [YOLOv5/YOLOv6/YOLOX/RTMDet 部署](https://github.com/open-mmlab/mmyolo/blob/main/configs/deploy)
82-
2. 支持 [YOLOv6](https://github.com/open-mmlab/mmyolo/blob/main/configs/yolov6) s/t/n 模型训练
83-
3. YOLOv5 支持 [P6 大分辨率 1280 尺度训练](https://github.com/open-mmlab/mmyolo/blob/main/configs/yolov5)
84-
4. YOLOv5 支持 [VOC 数据集训练](https://github.com/open-mmlab/mmyolo/blob/main/configs/yolov5/voc)
85-
5. 支持 [PPYOLOE](https://github.com/open-mmlab/mmyolo/blob/main/configs/ppyoloe) 和 [YOLOv7](https://github.com/open-mmlab/mmyolo/blob/main/configs/yolov7) 模型推理和官方权重转化
86-
6. How-to 文档中新增 YOLOv5 替换 [backbone 教程](https://github.com/open-mmlab/mmyolo/blob/dev/docs/zh_cn/advanced_guides/how_to.md#%E8%B7%A8%E5%BA%93%E4%BD%BF%E7%94%A8%E4%B8%BB%E5%B9%B2%E7%BD%91%E7%BB%9C)
76+
1. 支持 [YOLOv7](https://github.com/open-mmlab/mmyolo/tree/dev/configs/yolov7) P5 和 P6 模型
77+
2. 支持 [YOLOv6](https://github.com/open-mmlab/mmyolo/blob/dev/configs/yolov6/README.md) 中的 ML 大模型
78+
3. 支持 [Grad-Based CAM 和 Grad-Free CAM](https://github.com/open-mmlab/mmyolo/blob/dev/demo/boxam_vis_demo.py)
79+
4. 基于 sahi 支持 [大图推理](https://github.com/open-mmlab/mmyolo/blob/dev/demo/large_image_demo.py)
80+
5. projects 文件夹下新增 [easydeploy](https://github.com/open-mmlab/mmyolo/blob/dev/projects/easydeploy/README.md) 项目
81+
6. 新增 [自定义数据集教程](https://github.com/open-mmlab/mmyolo/blob/dev/docs/zh_cn/user_guides/custom_dataset.md)
8782

8883
同时我们也推出了解读视频：
8984

@@ -92,7 +87,9 @@ P6 模型图详见 [model_design.md](docs/zh_CN/algorithm_descriptions/model_des
9287
| 🌟 | 特征图可视化 | [![Link](https://i2.hdslb.com/bfs/archive/480a0eb41fce26e0acb65f82a74501418eee1032.jpg@112w_63h_1c.webp)](https://www.bilibili.com/video/BV188411s7o8) [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV188411s7o8)](https://www.bilibili.com/video/BV188411s7o8) | [特征图可视化.ipynb](https://github.com/open-mmlab/OpenMMLabCourse/blob/main/codes/MMYOLO_tutorials/%5B%E5%B7%A5%E5%85%B7%E7%B1%BB%E7%AC%AC%E4%B8%80%E6%9C%9F%5D%E7%89%B9%E5%BE%81%E5%9B%BE%E5%8F%AF%E8%A7%86%E5%8C%96.ipynb) |
9388
| 🌟 | 特征图可视化 Demo | [![Link](http://i0.hdslb.com/bfs/archive/081f300c84d6556f40d984cfbe801fc0644ff449.jpg@112w_63h_1c.webp)](https://www.bilibili.com/video/BV1je4y1478R/) [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1je4y1478R)](https://www.bilibili.com/video/BV1je4y1478R/) | |
9489
| 🌟 | 配置全解读 | [![Link](http://i1.hdslb.com/bfs/archive/e06daf640ea39b3c0700bb4dc758f1a253f33e13.jpg@112w_63h_1c.webp)](https://www.bilibili.com/video/BV1214y157ck) [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1214y157ck)](https://www.bilibili.com/video/BV1214y157ck) | [配置全解读文档](https://zhuanlan.zhihu.com/p/577715188) |
95-
| 🌟 | 源码阅读和调试「必备」技巧 | [![Link](https://i2.hdslb.com/bfs/archive/790d2422c879ff20488910da1c4422b667ea6af7.jpg@112w_63h_1c.webp)](https://www.bilibili.com/video/BV1N14y1V7mB) [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1N14y1V7mB)](https://www.bilibili.com/video/BV1N14y1V7mB) | |
90+
| 🌟 | 源码阅读和调试「必备」技巧 | [![Link](https://i2.hdslb.com/bfs/archive/790d2422c879ff20488910da1c4422b667ea6af7.jpg@112w_63h_1c.webp)](https://www.bilibili.com/video/BV1N14y1V7mB) [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1N14y1V7mB)](https://www.bilibili.com/video/BV1N14y1V7mB) | [源码阅读和调试「必备」技巧文档](https://zhuanlan.zhihu.com/p/580885852) |
91+
| 🌟 | 工程文件结构简析 | [![Link](http://i2.hdslb.com/bfs/archive/41030efb84d0cada06d5451c1e6e9bccc0cdb5a3.jpg@112w_63h_1c.webp)](https://www.bilibili.com/video/BV1LP4y117jS)[![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1LP4y117jS)](https://www.bilibili.com/video/BV1LP4y117jS) | [工程文件结构简析文档](https://zhuanlan.zhihu.com/p/584807195) |
92+
| 🌟 | 10分钟换遍主干网络 | [![Link](http://i0.hdslb.com/bfs/archive/c51f1aef7c605856777249a7b4478f44bd69f3bd.jpg@112w_63h_1c.webp)](https://www.bilibili.com/video/BV1JG4y1d7GC) [![bilibili](https://img.shields.io/badge/dynamic/json?label=views&style=social&logo=bilibili&query=data.stat.view&url=https%3A%2F%2Fapi.bilibili.com%2Fx%2Fweb-interface%2Fview%3Fbvid%3DBV1JG4y1d7GC)](https://www.bilibili.com/video/BV1JG4y1d7GC) | [10分钟换遍主干网络文档](https://zhuanlan.zhihu.com/p/585641598)<br>[10分钟换遍主干网络.ipynb](https://github.com/open-mmlab/OpenMMLabCourse/blob/main/codes/MMYOLO_tutorials/[实用类第二期]10分钟换遍主干网络.ipynb) |
9693

9794
发布历史和更新细节请参考 [更新日志](https://mmyolo.readthedocs.io/zh_CN/latest/notes/changelog.html)
9895

@@ -132,13 +129,15 @@ MMYOLO 用法和 MMDetection 几乎一致，所有教程都是通用的，你也
132129
- [实用工具](https://mmyolo.readthedocs.io/zh_CN/latest/user_guides/index.html#实用工具)
133130
- [可视化教程](docs/zh_cn/user_guides/visualization.md)
134131
- [实用工具](docs/zh_cn/user_guides/useful_tools.md)
132+
- [自定义数据集](docs/zh_cn/user_guides/custom_dataset.md)
135133

136134
- 算法描述
137135

138136
- [必备基础](https://mmyolo.readthedocs.io/zh_CN/latest/algorithm_descriptions/index.html#基础内容)
139137
- [模型设计相关说明](docs/zh_cn/algorithm_descriptions/model_design.md)
140138
- [算法原理和实现全解析](https://mmyolo.readthedocs.io/zh_CN/latest/algorithm_descriptions/index.html#算法原理和实现全解析)
141139
- [YOLOv5 原理和实现全解析](docs/zh_cn/algorithm_descriptions/yolov5_description.md)
140+
- [YOLOv6 原理和实现全解析](docs/zh_cn/algorithm_descriptions/yolov6_description.md)
142141
- [RTMDet 原理和实现全解析](docs/zh_cn/algorithm_descriptions/rtmdet_description.md)
143142

144143
- 算法部署
@@ -167,8 +166,8 @@ MMYOLO 用法和 MMDetection 几乎一致，所有教程都是通用的，你也
167166
- [x] [YOLOX](configs/yolox)
168167
- [x] [RTMDet](configs/rtmdet)
169168
- [x] [YOLOv6](configs/yolov6)
169+
- [x] [YOLOv7](configs/yolov7)
170170
- [ ] [PPYOLOE](configs/ppyoloe)(仅推理)
171-
- [ ] [YOLOv7](configs/yolov7)(仅推理)
172171

173172
</details>
174173

‎configs/deploy/detection_rknn-fp16_static-320x320.py

+9
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
_base_ = ['./base_static.py']
2+
onnx_config = dict(
3+
input_shape=[320, 320], output_names=['feat0', 'feat1', 'feat2'])
4+
codebase_config = dict(model_type='rknn')
5+
backend_config = dict(
6+
type='rknn',
7+
common_config=dict(target_platform='rv1126', optimization_level=1),
8+
quantization_config=dict(do_quantization=False, dataset=None),
9+
input_size_list=[[3, 320, 320]])

‎configs/deploy/detection_rknn-int8_static-320x320.py

+9
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,9 @@
1+
_base_ = ['./base_static.py']
2+
onnx_config = dict(
3+
input_shape=[320, 320], output_names=['feat0', 'feat1', 'feat2'])
4+
codebase_config = dict(model_type='rknn')
5+
backend_config = dict(
6+
type='rknn',
7+
common_config=dict(target_platform='rv1126', optimization_level=1),
8+
quantization_config=dict(do_quantization=True, dataset=None),
9+
input_size_list=[[3, 320, 320]])

‎configs/deploy/detection_tensorrt-int8_dynamic-192x192-960x960.py

+2-2
Original file line numberDiff line numberDiff line change
@@ -10,6 +10,6 @@
1010
min_shape=[1, 3, 192, 192],
1111
opt_shape=[1, 3, 640, 640],
1212
max_shape=[1, 3, 960, 960])))
13-
],
14-
calib_config=dict(create_calib=True, calib_file='calib_data.h5'))
13+
])
14+
calib_config = dict(create_calib=True, calib_file='calib_data.h5')
1515
use_efficientnms = False # whether to replace TRTBatchedNMS plugin with EfficientNMS plugin # noqa E501

‎configs/deploy/detection_tensorrt-int8_static-640x640.py

+2-2
Original file line numberDiff line numberDiff line change
@@ -11,6 +11,6 @@
1111
min_shape=[1, 3, 640, 640],
1212
opt_shape=[1, 3, 640, 640],
1313
max_shape=[1, 3, 640, 640])))
14-
],
15-
calib_config=dict(create_calib=True, calib_file='calib_data.h5'))
14+
])
15+
calib_config = dict(create_calib=True, calib_file='calib_data.h5')
1616
use_efficientnms = False # whether to replace TRTBatchedNMS plugin with EfficientNMS plugin # noqa E501

‎configs/yolov5/yolov5_s-v61_syncbn_8xb16-300e_coco.py

+4-1
Original file line numberDiff line numberDiff line change
@@ -19,6 +19,9 @@
1919
# persistent_workers must be False if num_workers is 0.
2020
persistent_workers = True
2121

22+
# Base learning rate for optim_wrapper
23+
base_lr = 0.01
24+
2225
# only on Val
2326
batch_shapes_cfg = dict(
2427
type='BatchShapePolicy',
@@ -199,7 +202,7 @@
199202
type='OptimWrapper',
200203
optimizer=dict(
201204
type='SGD',
202-
lr=0.01,
205+
lr=base_lr,
203206
momentum=0.937,
204207
weight_decay=0.0005,
205208
nesterov=True,

‎configs/yolov6/README.md

+11-8
Original file line numberDiff line numberDiff line change
@@ -16,17 +16,20 @@ For years, YOLO series have been de facto industry-level standard for efficient
1616

1717
### COCO
1818

19-
| Backbone | Arch | size | SyncBN | AMP | Mem (GB) | box AP | Config | Download |
20-
| :------: | :--: | :--: | :----: | :-: | :------: | :----: | :---------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
21-
| YOLOv6-n | P5 | 640 | Yes | Yes | 6.04 | 36.2 | [config](../yolov6/yolov6_n_syncbn_fast_8xb32-400e_coco.py) | [model](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_n_syncbn_fast_8xb32-400e_coco/yolov6_n_syncbn_fast_8xb32-400e_coco_20221030_202726-d99b2e82.pth) \| [log](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_n_syncbn_fast_8xb32-400e_coco/yolov6_n_syncbn_fast_8xb32-400e_coco_20221030_202726.log.json) |
22-
| YOLOv6-t | P5 | 640 | Yes | Yes | 8.13 | 41.0 | [config](../yolov6/yolov6_t_syncbn_fast_8xb32-400e_coco.py) | [model](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_t_syncbn_fast_8xb32-400e_coco/yolov6_t_syncbn_fast_8xb32-400e_coco_20221030_143755-cf0d278f.pth) \| [log](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_t_syncbn_fast_8xb32-400e_coco/yolov6_t_syncbn_fast_8xb32-400e_coco_20221030_143755.log.json) |
23-
| YOLOv6-s | P5 | 640 | Yes | Yes | 8.88 | 44.0 | [config](../yolov6/yolov6_s_syncbn_fast_8xb32-400e_coco.py) | [model](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_s_syncbn_fast_8xb32-400e_coco/yolov6_s_syncbn_fast_8xb32-400e_coco_20221102_203035-932e1d91.pth) \| [log](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_s_syncbn_fast_8xb32-400e_coco/yolov6_s_syncbn_fast_8xb32-400e_coco_20221102_203035.log.json) |
19+
| Backbone | Arch | Size | Epoch | SyncBN | AMP | Mem (GB) | Box AP | Config | Download |
20+
| :------: | :--: | :--: | :---: | :----: | :-: | :------: | :----: | :---------------------------------------------------------: | :--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------: |
21+
| YOLOv6-n | P5 | 640 | 400 | Yes | Yes | 6.04 | 36.2 | [config](../yolov6/yolov6_n_syncbn_fast_8xb32-400e_coco.py) | [model](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_n_syncbn_fast_8xb32-400e_coco/yolov6_n_syncbn_fast_8xb32-400e_coco_20221030_202726-d99b2e82.pth) \| [log](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_n_syncbn_fast_8xb32-400e_coco/yolov6_n_syncbn_fast_8xb32-400e_coco_20221030_202726.log.json) |
22+
| YOLOv6-t | P5 | 640 | 400 | Yes | Yes | 8.13 | 41.0 | [config](../yolov6/yolov6_t_syncbn_fast_8xb32-400e_coco.py) | [model](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_t_syncbn_fast_8xb32-400e_coco/yolov6_t_syncbn_fast_8xb32-400e_coco_20221030_143755-cf0d278f.pth) \| [log](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_t_syncbn_fast_8xb32-400e_coco/yolov6_t_syncbn_fast_8xb32-400e_coco_20221030_143755.log.json) |
23+
| YOLOv6-s | P5 | 640 | 400 | Yes | Yes | 8.88 | 44.0 | [config](../yolov6/yolov6_s_syncbn_fast_8xb32-400e_coco.py) | [model](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_s_syncbn_fast_8xb32-400e_coco/yolov6_s_syncbn_fast_8xb32-400e_coco_20221102_203035-932e1d91.pth) \| [log](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_s_syncbn_fast_8xb32-400e_coco/yolov6_s_syncbn_fast_8xb32-400e_coco_20221102_203035.log.json) |
24+
| YOLOv6-m | P5 | 640 | 300 | Yes | Yes | 16.69 | 48.4 | [config](../yolov6/yolov6_m_syncbn_fast_8xb32-400e_coco.py) | [model](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_m_syncbn_fast_8xb32-300e_coco/yolov6_m_syncbn_fast_8xb32-300e_coco_20221109_182658-85bda3f4.pth) \| [log](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_m_syncbn_fast_8xb32-300e_coco/yolov6_m_syncbn_fast_8xb32-300e_coco_20221109_182658.log.json) |
25+
| YOLOv6-l | P5 | 640 | 300 | Yes | Yes | 20.86 | 51.0 | [config](../yolov6/yolov6_l_syncbn_fast_8xb32-300e_coco.py) | [model](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_l_syncbn_fast_8xb32-300e_coco/yolov6_l_syncbn_fast_8xb32-300e_coco_20221109_183156-91e3c447.pth) \| [log](https://download.openmmlab.com/mmyolo/v0/yolov6/yolov6_l_syncbn_fast_8xb32-300e_coco/yolov6_l_syncbn_fast_8xb32-300e_coco_20221109_183156.log.json) |
2426

2527
**Note**:
2628

27-
1. The performance is unstable and may fluctuate by about 0.3 mAP.
28-
2. YOLOv6-m,l,x will be supported in later version.
29-
3. If users need the weight of 300 epoch, they can train according to the configs of 300 epoch provided by us, or convert the official weight according to the [converter script](../../tools/model_converters/).
29+
1. The official m and l models use knowledge distillation, but our version does not support it, which will be implemented in [MMRazor](https://github.com/open-mmlab/mmrazor) in the future.
30+
2. The performance is unstable and may fluctuate by about 0.3 mAP.
31+
3. If users need the weight of 300 epoch for nano, tiny and small model, they can train according to the configs of 300 epoch provided by us, or convert the official weight according to the [converter script](../../tools/model_converters/).
32+
4. We have observed that the [base model](https://github.com/meituan/YOLOv6/tree/main/configs/base) has been officially released in v6 recently. Although the accuracy has decreased, it is more efficient. We will also provide the base model configuration in the future.
3033

3134
## Citation
3235

0 commit comments

Comments
 (0)