Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Docs] Add Chinese version of finetune #7178

Merged
merged 9 commits into from
Mar 3, 2022
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 7 additions & 4 deletions docs/en/get_started.md
Original file line number Diff line number Diff line change
Expand Up @@ -134,7 +134,7 @@ Or you can still install MMDetection manually:
# for LVIS dataset
pip install git+https://github.com/lvis-dataset/lvis-api.git
# for albumentations
pip install albumentations>=0.3.2 --no-binary imgaug,albumentations
pip install -r requirements/albu.txt
```

**Note:**
Expand All @@ -148,9 +148,12 @@ you can install it before installing MMCV.
c. Some dependencies are optional. Simply running `pip install -v -e .` will
only install the minimum runtime requirements. To use optional dependencies like `albumentations` and `imagecorruptions` either install them manually with `pip install -r requirements/optional.txt` or specify desired extras when calling `pip` (e.g. `pip install -v -e .[optional]`). Valid keys for the extras field are: `all`, `tests`, `build`, and `optional`.

d. If you would like to use `albumentations`, we suggest using
`pip install albumentations>=0.3.2 --no-binary imgaug,albumentations`. If you simply use
`pip install albumentations>=0.3.2`, it will install `opencv-python-headless` simultaneously (even though you have already installed `opencv-python`). We should not allow `opencv-python` and `opencv-python-headless` installed at the same time, because it might cause unexpected issues. Please refer to [official documentation](https://albumentations.ai/docs/getting_started/installation/#note-on-opencv-dependencies) for more details.
d. If you would like to use `albumentations`, we suggest using `pip install -r requirements/albu.txt` or
`pip install -U albumentations --no-binary qudida,albumentations`. If you simply use `pip install albumentations>=0.3.2`,
it will install `opencv-python-headless` simultaneously (even though you have already
installed `opencv-python`). We recommended checking the environment after installing `albumentation` to
ensure that `opencv-python` and `opencv-python-headless` are not installed at the same time, because it might cause unexpected issues if they both installed. Please refer
to [official documentation](https://albumentations.ai/docs/getting_started/installation/#note-on-opencv-dependencies) for more details.

### Install without GPU support

Expand Down
4 changes: 2 additions & 2 deletions docs/zh_cn/get_started.md
Original file line number Diff line number Diff line change
Expand Up @@ -142,7 +142,7 @@ MIM 能够自动地安装 OpenMMLab 的项目以及对应的依赖包。
# 安装 LVIS 数据集依赖
pip install git+https://github.com/lvis-dataset/lvis-api.git
# 安装 albumentations 依赖
pip install albumentations>=0.3.2 --no-binary imgaug,albumentations
pip install -r requirements/albu.txt
```

**注意:**
Expand All @@ -153,7 +153,7 @@ MIM 能够自动地安装 OpenMMLab 的项目以及对应的依赖包。

(3) 一些安装依赖是可以选择的。例如只需要安装最低运行要求的版本,则可以使用 `pip install -v -e .` 命令。如果希望使用可选择的像 `albumentations` 和 `imagecorruptions` 这种依赖项,可以使用 `pip install -r requirements/optional.txt` 进行手动安装,或者在使用 `pip` 时指定所需的附加功能(例如 `pip install -v -e .[optional]`),支持附加功能的有效键值包括 `all`、`tests`、`build` 以及 `optional` 。

(4) 如果希望使用 `albumentations`,我们建议使用 `pip install albumentations>=0.3.2 --no-binary imgaug,albumentations` 进行安装。 如果简单地使用 `pip install albumentations>=0.3.2` 进行安装,则会同时安装 `opencv-python-headless`(即便已经安装了 `opencv-python` 也会再次安装)。我们不允许同时安装 `opencv-python` 和 `opencv-python-headless`,因为这样可能会导致一些问题。更多细节请参考[官方文档](https://albumentations.ai/docs/getting_started/installation/#note-on-opencv-dependencies)。
(4) 如果希望使用 `albumentations`,我们建议使用 `pip install -r requirements/albu.txt` 或者 `pip install -U albumentations --no-binary qudida,albumentations` 进行安装。 如果简单地使用 `pip install albumentations>=0.3.2` 进行安装,则会同时安装 `opencv-python-headless`(即便已经安装了 `opencv-python` 也会再次安装)。我们建议在安装 `albumentations` 后检查环境,以确保没有同时安装 `opencv-python` 和 `opencv-python-headless`,因为同时安装可能会导致一些问题。更多细节请参考[官方文档](https://albumentations.ai/docs/getting_started/installation/#note-on-opencv-dependencies)。

### 只在 CPU 安装

Expand Down
83 changes: 83 additions & 0 deletions docs/zh_cn/tutorials/finetune.md
Original file line number Diff line number Diff line change
@@ -1 +1,84 @@
# 教程 7: 模型微调

在 COCO 数据集上预训练的检测器可以作为其他数据集(例如 CityScapes 和 KITTI 数据集)优质的预训练模型。
本教程将指导用户如何把 [ModelZoo](../model_zoo.md) 中提供的模型用于其他数据集中并使得当前所训练的模型获得更好性能。

以下是在新数据集中微调模型需要的两个步骤。

- 按 [教程2:自定义数据集的方法](customize_dataset.md) 中的方法对新数据集添加支持中的方法对新数据集添加支持
- 按照本教程中所讨论方法,修改配置信息

接下来将会以 Cityscapes Dataset 上的微调过程作为例子,具体讲述用户需要在配置中修改的五个部分。

## 继承基础配置

为了减轻编写整个配置的负担并减少漏洞的数量, MMDetection V2.0 支持从多个现有配置中继承配置信息。微调 MaskRCNN 模型的时候,新的配置信息需要使用从 `_base_/models/mask_rcnn_r50_fpn.py`中继承的配置信息来构建模型的基本结构。当使用 Cityscapes 数据集时,新的配置信息可以简便地从`_base_/datasets/cityscapes_instance.py`中继承。对于训练过程的运行设置部分,新配置需要从 `_base_/default_runtime.py`中继承。这些配置文件`configs`的目录下,用户可以选择全部内容的重新编写而不是使用继承方法。

```python
_base_ = [
'../_base_/models/mask_rcnn_r50_fpn.py',
'../_base_/datasets/cityscapes_instance.py', '../_base_/default_runtime.py'
]
```


## Head 的修改
接下来新的配置还需要根据新数据集的类别数量对 Head 进行修改。只需要对 roi_head 中的 `num_classes`进行修改。修改后除了最后的预测模型的 Head 之外,预训练模型的权重的大部分都会被重新使用。

```python
model = dict(
pretrained=None,
roi_head=dict(
bbox_head=dict(
type='Shared2FCBBoxHead',
in_channels=256,
fc_out_channels=1024,
roi_feat_size=7,
num_classes=8,
bbox_coder=dict(
type='DeltaXYWHBBoxCoder',
target_means=[0., 0., 0., 0.],
target_stds=[0.1, 0.1, 0.2, 0.2]),
reg_class_agnostic=False,
loss_cls=dict(
type='CrossEntropyLoss', use_sigmoid=False, loss_weight=1.0),
loss_bbox=dict(type='SmoothL1Loss', beta=1.0, loss_weight=1.0)),
mask_head=dict(
type='FCNMaskHead',
num_convs=4,
in_channels=256,
conv_out_channels=256,
num_classes=8,
loss_mask=dict(
type='CrossEntropyLoss', use_mask=True, loss_weight=1.0))))
```

## 数据集的修改
用户可能还需要准备数据集并编写有关数据集的配置。目前 MMDetection V2.0 的配置文件已经支持 VOC、WIDER FACE、COCO 和 Cityscapes Dataset 的数据集信息。

## 训练策略的修改
微调超参数与默认的训练策略不同。它通常需要更小的学习率和更少的训练回合。

```python
# 优化器
# batch size 为 8 时的 lr 配置
optimizer = dict(type='SGD', lr=0.01, momentum=0.9, weight_decay=0.0001)
optimizer_config = dict(grad_clip=None)
# 学习策略
lr_config = dict(
policy='step',
warmup='linear',
warmup_iters=500,
warmup_ratio=0.001,
step=[7])
# lr_config 中的 max_epochs 和 step 需要针对自定义数据集进行专门调整
runner = dict(max_epochs=8)
log_config = dict(interval=100)
```

## 使用预训练模型

如果要使用预训练模型时,可以在 `load_from` 中查阅新的配置信息,用户需要在训练开始之前下载好需要的模型权重,从而避免在训练过程中浪费了宝贵时间。
```python
load_from = 'https://download.openmmlab.com/mmdetection/v2.0/mask_rcnn/mask_rcnn_r50_caffe_fpn_mstrain-poly_3x_coco/mask_rcnn_r50_caffe_fpn_mstrain-poly_3x_coco_bbox_mAP-0.408__segm_mAP-0.37_20200504_163245-42aa3d00.pth' # noqa
```
1 change: 1 addition & 0 deletions requirements/albu.txt
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
albumentations>=0.3.2 --no-binary qudida,albumentations
2 changes: 1 addition & 1 deletion tools/deployment/onnx2tensorrt.py
Original file line number Diff line number Diff line change
Expand Up @@ -201,7 +201,7 @@ def parse_args():
parsed directly from config file and are deprecated and will be \
removed in future releases.')
if not args.input_img:
args.input_img = osp.join(osp.dirname(__file__), '../demo/demo.jpg')
args.input_img = osp.join(osp.dirname(__file__), '../../demo/demo.jpg')

cfg = Config.fromfile(args.config)

Expand Down