[Docs] Refine config.md

docs/en/user_guides/config.md

@@ -286,7 +286,7 @@ optim_wrapper = dict(  # Optimizer wrapper config
     constructor='YOLOv5OptimizerConstructor') # The constructor for YOLOv5 optimizer
 ```
 
-`param_scheduler` is the field that configures methods of adjusting optimization hyperparameters such as learning rate and momentum. Users can combine multiple schedulers to create a desired parameter adjustment strategy. Find more in the [parameter scheduler tutorial](https://mmengine.readthedocs.io/en/latest/tutorials/param_scheduler.html). In MMYOLO, no parameter optimizer is introduced.
+`param_scheduler` is the field that configures methods of adjusting optimization hyperparameters such as learning rate and momentum. Users can combine multiple schedulers to create a desired parameter adjustment strategy. Find more in the [parameter scheduler tutorial](https://mmengine.readthedocs.io/en/latest/tutorials/param_scheduler.html). In YOLOv5, parameter scheduling is complex to implement and difficult to implement with `param_scheduler`. So we use `YOLOv5ParamSchedulerHook` to implement it (see next section), which is simpler but less versatile.
 
 ```python
 param_scheduler = None
@@ -387,26 +387,29 @@ If you wish to inspect the config file, you may run `mim run mmdet print_config
 Sometimes, you may set `_delete_=True` to ignore some of the fields in base configs.
 You may refer to the [mmengine config tutorial](https://mmengine.readthedocs.io/en/latest/tutorials/config.html) for a simple illustration.
 
-In MMYOLO, for example, to change the backbone of YOLOv5 with the following config.
+In MMYOLO, for example, to change the backbone of RTMDet with the following config.
 
 ```python
 model = dict(
     type='YOLODetector',
     data_preprocessor=dict(...),
     backbone=dict(
-        type='YOLOv5CSPDarknet',
+        type='CSPNeXt',
+        arch='P5',
+        expand_ratio=0.5,
         deepen_factor=deepen_factor,
         widen_factor=widen_factor,
-        norm_cfg=dict(type='BN', momentum=0.03, eps=0.001),
+        channel_attention=True,
+        norm_cfg=dict(type='BN'),
         act_cfg=dict(type='SiLU', inplace=True)),
     neck=dict(...),
     bbox_head=dict(...))
 ```
 
-The `_delete_=True` would replace all old keys in the `backbone` field with new keys. For example, `YOLOv5` uses `YOLOv5CSPDarknet`, it is necessary to replace the backbone with `YOLOv6EfficientRep`. Since `YOLOv5CSPDarknet` and `YOLOv6EfficientRep` have different fields, you need to use `_delete_=True` to replace all old keys in the `backbone` field.
+If you want to change `CSPNeXt` to `YOLOv6EfficientRep` for the RTMDet backbone, because there are different fields (`channel_attention` and `expand_ratio`) in `CSPNeXt` and `YOLOv6EfficientRep`, you need to use `_delete_=True` to replace all the old keys in the `backbone` field with the new keys.
 
 ```python
-_base_ = '../yolov5/yolov5_s-v61_syncbn_8xb16-300e_coco.py'
+_base_ = '../rtmdet/rtmdet_l_syncbn_8xb32-300e_coco.py'
 model = dict(
     backbone=dict(
         _delete_=True,

docs/zh_cn/user_guides/config.md

@@ -287,7 +287,7 @@ optim_wrapper = dict(  # 优化器封装的配置
 
 ```
 
-`param_scheduler` 字段用于配置参数调度器（Parameter Scheduler）来调整优化器的超参数（例如学习率和动量）。 用户可以组合多个调度器来创建所需的参数调整策略。 在[参数调度器教程](https://mmengine.readthedocs.io/zh_CN/latest/tutorials/param_scheduler.html) 和参数调度器 API 文档 中查找更多信息。在 MMYOLO 中，未引入任何参数调度器。
+`param_scheduler` 字段用于配置参数调度器（Parameter Scheduler）来调整优化器的超参数（例如学习率和动量）。 用户可以组合多个调度器来创建所需的参数调整策略。 在[参数调度器教程](https://mmengine.readthedocs.io/zh_CN/latest/tutorials/param_scheduler.html) 和参数调度器 API 文档 中查找更多信息。在 YOLOv5 中，参数调度实现比较复杂，难以通过  `param_scheduler` 实现。所以我们采用了 `YOLOv5ParamSchedulerHook` 来实现（见下节），这样做更简单但是通用性较差。
 
 ```python
 param_scheduler = None
@@ -386,26 +386,29 @@ _base_ = [
 
 有时，您也许会设置 `_delete_=True` 去忽略基础配置文件里的一些域内容。 您也许可以参照 [MMEngine 配置文件教程](https://mmengine.readthedocs.io/en/latest/tutorials/config.html) 来获得一些简单的指导。
 
-在 MMYOLO 里，例如为了改变  YOLOv5 的主干网络的某些内容：
+在 MMYOLO 里，例如为了改变 RTMDet 的主干网络的某些内容：
 
 ```python
 model = dict(
     type='YOLODetector',
     data_preprocessor=dict(...),
     backbone=dict(
-        type='YOLOv5CSPDarknet',
+        type='CSPNeXt',
+        arch='P5',
+        expand_ratio=0.5,
         deepen_factor=deepen_factor,
         widen_factor=widen_factor,
-        norm_cfg=dict(type='BN', momentum=0.03, eps=0.001),
+        channel_attention=True,
+        norm_cfg=dict(type='BN'),
         act_cfg=dict(type='SiLU', inplace=True)),
     neck=dict(...),
     bbox_head=dict(...))
 ```
 
-基础配置的 `YOLOv5` 使用 `YOLOv5CSPDarknet`，在需要将主干网络改成 `YOLOv6EfficientRep` 的时候，因为 `YOLOv5CSPDarknet` 和 `YOLOv6EfficientRep` 中有不同的字段，需要使用 `_delete_=True` 将新的键去替换 `backbone` 域内所有老的键。
+如果想把 RTMDet 主干网络的 `CSPNeXt` 改成 `YOLOv6EfficientRep`，因为 `CSPNeXt` 和 `YOLOv6EfficientRep` 中有不同的字段(`channel_attention` 和 `expand_ratio`)，这时候就需要使用 `_delete_=True` 将新的键去替换 `backbone` 域内所有老的键。
 
 ```python
-_base_ = '../yolov5/yolov5_s-v61_syncbn_8xb16-300e_coco.py'
+_base_ = '../rtmdet/rtmdet_l_syncbn_8xb32-300e_coco.py'
 model = dict(
     backbone=dict(
         _delete_=True,

mmyolo/models/backbones/cspnext.py

@@ -28,12 +28,13 @@ class CSPNeXt(BaseBackbone):
         frozen_stages (int): Stages to be frozen (stop grad and set eval
             mode). -1 means not freezing any parameters. Defaults to -1.
         plugins (list[dict]): List of plugins for stages, each dict contains:
-
-            - cfg (dict, required): Cfg dict to build plugin.
+            - cfg (dict, required): Cfg dict to build plugin.Defaults to
             - stages (tuple[bool], optional): Stages to apply plugin, length
               should be same as 'num_stages'.
         use_depthwise (bool): Whether to use depthwise separable convolution.
             Defaults to False.
+        expand_ratio (float): Ratio to adjust the number of channels of the
+            hidden layer. Defaults to 0.5.
         arch_ovewrite (list): Overwrite default arch settings.
             Defaults to None.
         channel_attention (bool): Whether to add channel attention in each

mmyolo/models/necks/cspnext_pafpn.py

@@ -27,7 +27,7 @@ class CSPNeXtPAFPN(BaseYOLONeck):
         use_depthwise (bool): Whether to use depthwise separable convolution in
             blocks. Defaults to False.
         expand_ratio (float): Ratio to adjust the number of channels of the
-            hidden layer. Default: 0.5
+            hidden layer. Defaults to 0.5.
         upsample_cfg (dict): Config dict for interpolate layer.
             Default: `dict(scale_factor=2, mode='nearest')`
         conv_cfg (dict, optional): Config dict for convolution layer.