forked from mindspore-Ecosystem/mindspore
!17321 update README for SSD
From: @zhouneng2 Reviewed-by: @linqingke,@liangchenghui Signed-off-by: @linqingke
This commit is contained in:
Gitee
commit
ddf4b5cf88
|
|
@ -192,26 +192,40 @@ Then you can run everything just like on ascend.
|
|||
.
|
||||
└─ cv
|
||||
└─ ssd
|
||||
├─ README.md # descriptions about SSD
|
||||
├─ README.md ## descriptions about SSD
|
||||
├─ ascend310_infer ## source code of 310 inference
|
||||
├─ scripts
|
||||
├─ run_distribute_train.sh # shell script for distributed on ascend
|
||||
├─ run_distribute_train_gpu.sh # shell script for distributed on gpu
|
||||
├─ run_eval.sh # shell script for eval on ascend
|
||||
└─ run_eval_gpu.sh # shell script for eval on gpu
|
||||
├─ docker start.sh ## shell script for start docker container
|
||||
├─ run_distribute_train.sh ## shell script for distributed on ascend
|
||||
├─ run_distribute_train_gpu.sh ## shell script for distributed on gpu
|
||||
├─ run_eval.sh ## shell script for eval on ascend
|
||||
├─ run_eval_gpu.sh ## shell script for eval on gpu
|
||||
└─ run_infer_310.sh ## shell script for 310 inference
|
||||
├─ src
|
||||
├─ __init__.py # init file
|
||||
├─ box_utils.py # bbox utils
|
||||
├─ eval_callback.py # evaluation callback when training
|
||||
├─ eval_utils.py # metrics utils
|
||||
├─ config.py # total config
|
||||
├─ dataset.py # create dataset and process dataset
|
||||
├─ init_params.py # parameters utils
|
||||
├─ lr_schedule.py # learning ratio generator
|
||||
└─ ssd.py # ssd architecture
|
||||
├─ eval.py # eval scripts
|
||||
├─ train.py # train scripts
|
||||
├─ export.py # export mindir script
|
||||
└─ mindspore_hub_conf.py # mindspore hub interface
|
||||
├─ __init__.py ## init file
|
||||
├─ anchor_generator.py ## anchor generator
|
||||
├─ box_util.py ## bbox utils
|
||||
├─ config.py ## total config
|
||||
├─ config_ssd_mobilenet_v1_fpn.py ## configuration for feature extraction network uses mobilenet-v1 with fpn
|
||||
├─ config_ssd_resnet50_fpn.py ## configuration for feature extraction network uses resnet50 with fpn
|
||||
├─ config_ssd_vgg16.py ## configuration for backbone of vgg16
|
||||
├─ config_ssd300.py ## configuration for backbone of mobilenet-v2
|
||||
├─ dataset.py ## create dataset and process dataset
|
||||
├─ eval_callback.py ## eval callback function definition
|
||||
├─ eval_utils.py ## eval utils
|
||||
├─ fpn.py ## feature pyramid network
|
||||
├─ init_params.py ## parameters utils
|
||||
├─ lr_schedule.py ## learning ratio generator
|
||||
├─ mobilenet_v1.py ## network definition for mobilenet-v1
|
||||
├─ resnet.py ## network definition for resnet
|
||||
├─ ssd.py ## ssd architecture
|
||||
└─ vgg16.py ## network definition for vgg16
|
||||
├─ Dockerfile ## docker file
|
||||
├─ eval.py ## eval scripts
|
||||
├─ export.py ## export mindir script
|
||||
├─ postprocess.py ## post-processing script for 310 inference
|
||||
├─ train.py ## train scripts
|
||||
└─ mindspore_hub_conf.py ## mindspore hub interface
|
||||
```
|
||||
|
||||
### [Script Parameters](#contents)
|
||||
|
|
@ -295,7 +309,7 @@ epoch time: 39064.8467540741, per step time: 85.29442522723602
|
|||
bash run_distribute_train_gpu.sh [DEVICE_NUM] [EPOCH_SIZE] [LR] [DATASET] [PRE_TRAINED](optional) [PRE_TRAINED_EPOCH_SIZE](optional)
|
||||
```
|
||||
|
||||
We need five or seven parameters for this scripts.
|
||||
We need four or six parameters for this scripts.
|
||||
|
||||
- `DEVICE_NUM`: the device number for distributed train.
|
||||
- `EPOCH_NUM`: epoch num for distributed train.
|
||||
|
|
|
|||
|
|
@ -40,6 +40,13 @@ SSD将边界框的输出空间离散成一组默认框，每个特征映射位
|
|||
|
||||
SSD方法基于前向卷积网络，该网络产生固定大小的边界框集合，并针对这些框内存在的对象类实例进行评分，然后通过非极大值抑制步骤进行最终检测。早期的网络层基于高质量图像分类的标准体系结构，被称为基础网络。后来通过向网络添加辅助结构进行检测。
|
||||
|
||||
我们提供了4种不同的基础架构:
|
||||
|
||||
- **ssd300**, 参考论文实现. 使用mobilenet-v2作为骨干网络, 并使用相同的bbox预测器.
|
||||
- ***ssd-mobilenet-v1-fpn**, 使用mobilenet-v1和FPN作为特征提取器, 并使用权重共享bbox预测器.
|
||||
- ***ssd-resnet50-fpn**, 使用resnet50和FPN作为特征提取器, 并使用权重共享bbox预测器.
|
||||
- **ssd-vgg16**, 参考论文实现. 使用vgg16作为骨干网络, 并使用相同的bbox预测器.
|
||||
|
||||
# 数据集
|
||||
|
||||
使用的数据集： [COCO2017](<http://images.cocodataset.org/>)
|
||||
|
|
@ -129,23 +136,39 @@ sh run_eval_gpu.sh [DATASET] [CHECKPOINT_PATH] [DEVICE_ID]
|
|||
└─ cv
|
||||
└─ ssd
|
||||
├─ README.md ## SSD相关说明
|
||||
├─ ascend310_infer ## 实现310推理源代码
|
||||
├─ scripts
|
||||
├─ docker start.sh ## 容器启动脚本
|
||||
├─ run_distribute_train.sh ## Ascend分布式shell脚本
|
||||
├─ run_distribute_train_gpu.sh ## GPU分布式shell脚本
|
||||
├─ run_eval.sh ## Ascend评估shell脚本
|
||||
└─ run_eval_gpu.sh ## GPU评估shell脚本
|
||||
├─ run_eval_gpu.sh ## GPU评估shell脚本
|
||||
└─ run_infer_310.sh ## 310推理脚本
|
||||
├─ src
|
||||
├─ __init__.py ## 初始化文件
|
||||
├─ box_util.py ## bbox工具
|
||||
├─ coco_eval.py ## coco指标工具
|
||||
├─ config.py ## 总配置
|
||||
├─ dataset.py ## 创建并处理数据集
|
||||
├─ init_params.py ## 参数工具
|
||||
├─ lr_schedule.py ## 学习率生成器
|
||||
└─ ssd.py ## SSD架构
|
||||
├─ eval.py ## 评估脚本
|
||||
├─ train.py ## 训练脚本
|
||||
└─ mindspore_hub_conf.py ## MindSpore Hub接口
|
||||
├─ __init__.py ## 初始化文件
|
||||
├─ anchor_generator.py ## 锚点生成器
|
||||
├─ box_util.py ## bbox工具
|
||||
├─ config.py ## 总配置
|
||||
├─ config_ssd_mobilenet_v1_fpn.py ## 特征提取网络使用mobilenet-v1 with fpn配置
|
||||
├─ config_ssd_resnet50_fpn.py ## 特征提取网络使用resnet50 with fpn 配置
|
||||
├─ config_ssd_vgg16.py ## 特征提取网络使用vgg16 配置
|
||||
├─ config_ssd300.py ## 特征提取网络使用mobilenet-v2 配置
|
||||
├─ dataset.py ## 创建并处理数据集
|
||||
├─ eval_callback.py ## eval回调方法定义
|
||||
├─ eval_utils.py ## eval工具
|
||||
├─ fpn.py ## 特征金字塔网络
|
||||
├─ init_params.py ## 参数工具
|
||||
├─ lr_schedule.py ## 学习率生成器
|
||||
├─ mobilenet_v1.py ## mobilenet-v1网络定义
|
||||
├─ resnet.py ## resnet网络定义
|
||||
├─ ssd.py ## SSD架构
|
||||
└─ vgg16.py ## vgg16网络定义
|
||||
├─ Dockerfile ## docker文件
|
||||
├─ eval.py ## 评估脚本
|
||||
├─ export.py ## 导出 AIR,MINDIR模型的脚本
|
||||
├─ postprocess.py ## 310推理后处理脚本
|
||||
├─ train.py ## 训练脚本
|
||||
└─ mindspore_hub_conf.py ## MindSpore Hub接口
|
||||
```
|
||||
|
||||
## 脚本参数
|
||||
|
|
@ -222,7 +245,7 @@ epoch time: 39064.8467540741, per step time: 85.29442522723602
|
|||
sh run_distribute_train_gpu.sh [DEVICE_NUM] [EPOCH_SIZE] [LR] [DATASET] [PRE_TRAINED](optional) [PRE_TRAINED_EPOCH_SIZE](optional)
|
||||
```
|
||||
|
||||
此脚本需要五或七个参数。
|
||||
此脚本需要四或六个参数。
|
||||
|
||||
- `DEVICE_NUM`：分布式训练的设备数。
|
||||
- `EPOCH_NUM`：分布式训练的轮次数。
|
||||
|
|
|
|||
Loading…
Reference in New Issue