未验证 提交 f773e869 编写于 作者: W wangguanzhong 提交者: GitHub

cherry pick polish doc (#329)

上级 feef394d
...@@ -160,6 +160,7 @@ improved performance mainly by using L1 loss in bounding box width and height re ...@@ -160,6 +160,7 @@ improved performance mainly by using L1 loss in bounding box width and height re
randomly color distortion, randomly cropping, randomly expansion, randomly interpolation method, randomly flippling. YOLO v3 used randomly randomly color distortion, randomly cropping, randomly expansion, randomly interpolation method, randomly flippling. YOLO v3 used randomly
reshaped minibatch in training, inferences can be performed on different image sizes with the same model weights, and we provided evaluation reshaped minibatch in training, inferences can be performed on different image sizes with the same model weights, and we provided evaluation
results of image size 608/416/320 above. Deformable conv is added on stage 5 of backbone. results of image size 608/416/320 above. Deformable conv is added on stage 5 of backbone.
- YOLO v3 enhanced model improves the precision to 43.2 involved with deformable conv, dropblock and IoU loss. See more details in [YOLOv3_ENHANCEMENT](./featured_model/YOLOv3_ENHANCEMENT.md)
### RetinaNet ### RetinaNet
......
...@@ -153,6 +153,7 @@ Paddle提供基于ImageNet的骨架网络预训练模型。所有预训练模型 ...@@ -153,6 +153,7 @@ Paddle提供基于ImageNet的骨架网络预训练模型。所有预训练模型
- 上表中也提供了原论文[YOLOv3](https://arxiv.org/abs/1804.02767)中YOLOv3-DarkNet53的精度,我们的实现版本主要从在bounding box的宽度和高度回归上使用了L1损失,图像mixup和label smooth等方法优化了其精度。 - 上表中也提供了原论文[YOLOv3](https://arxiv.org/abs/1804.02767)中YOLOv3-DarkNet53的精度,我们的实现版本主要从在bounding box的宽度和高度回归上使用了L1损失,图像mixup和label smooth等方法优化了其精度。
- YOLO v3在8卡,总batch size为64下训练270轮。数据增强包括:mixup, 随机颜色失真,随机剪裁,随机扩张,随机插值法,随机翻转。YOLO v3在训练阶段对minibatch采用随机reshape,可以采用相同的模型测试不同尺寸图片,我们分别提供了尺寸为608/416/320大小的测试结果。deformable卷积作用在骨架网络5阶段。 - YOLO v3在8卡,总batch size为64下训练270轮。数据增强包括:mixup, 随机颜色失真,随机剪裁,随机扩张,随机插值法,随机翻转。YOLO v3在训练阶段对minibatch采用随机reshape,可以采用相同的模型测试不同尺寸图片,我们分别提供了尺寸为608/416/320大小的测试结果。deformable卷积作用在骨架网络5阶段。
- YOLO v3增强版模型通过引入可变形卷积,dropblock和IoU loss,将精度进一步提升至43.2, 详情见[YOLOv3增强模型](./featured_model/YOLOv3_ENHANCEMENT.md)
### RetinaNet ### RetinaNet
......
...@@ -184,7 +184,7 @@ A small utility (`tools/configure.py`) is included to simplify the configuration ...@@ -184,7 +184,7 @@ A small utility (`tools/configure.py`) is included to simplify the configuration
4. `generate`: Generate a configuration template for a given list of modules. By default it generates a complete configuration file, which can be quite verbose; if a `--minimal` flag is given, it generates a template that only contain non optional settings. For example, to generate a configuration for Faster R-CNN architecture with `ResNet` backbone and `FPN`, run: 4. `generate`: Generate a configuration template for a given list of modules. By default it generates a complete configuration file, which can be quite verbose; if a `--minimal` flag is given, it generates a template that only contain non optional settings. For example, to generate a configuration for Faster R-CNN architecture with `ResNet` backbone and `FPN`, run:
```shell ```shell
python tools/configure.py generate FasterRCNN ResNet RPNHead RoIAlign BBoxAssigner BBoxHead FasterRCNNTrainFeed FasterRCNNTestFeed LearningRate OptimizerBuilder python tools/configure.py generate FasterRCNN ResNet RPNHead RoIAlign BBoxAssigner BBoxHead LearningRate OptimizerBuilder
``` ```
For a minimal version, run: For a minimal version, run:
......
...@@ -174,7 +174,7 @@ pip install typeguard http://github.com/willthefrog/docstring_parser/tarball/mas ...@@ -174,7 +174,7 @@ pip install typeguard http://github.com/willthefrog/docstring_parser/tarball/mas
4. `generate`: 根据给出的模块列表生成配置文件,默认生成完整配置,如果指定 `--minimal` ,生成最小配置,即省略所有默认配置项。例如,执行下列命令可以生成Faster R-CNN (`ResNet` backbone + `FPN`) 架构的配置文件: 4. `generate`: 根据给出的模块列表生成配置文件,默认生成完整配置,如果指定 `--minimal` ,生成最小配置,即省略所有默认配置项。例如,执行下列命令可以生成Faster R-CNN (`ResNet` backbone + `FPN`) 架构的配置文件:
```shell ```shell
python tools/configure.py generate FasterRCNN ResNet RPNHead RoIAlign BBoxAssigner BBoxHead FasterRCNNTrainFeed FasterRCNNTestFeed LearningRate OptimizerBuilder python tools/configure.py generate FasterRCNN ResNet RPNHead RoIAlign BBoxAssigner BBoxHead LearningRate OptimizerBuilder
``` ```
如需最小配置,运行: 如需最小配置,运行:
......
...@@ -2,7 +2,7 @@ ...@@ -2,7 +2,7 @@
本篇教程使用Python API对[导出模型](EXPORT_MODEL.md)保存的inference_model进行预测。 本篇教程使用Python API对[导出模型](EXPORT_MODEL.md)保存的inference_model进行预测。
在PaddlePaddle中预测引擎和训练引擎底层有着不同的优化方法,代码走不的分支,两者都可以进行预测。在入门教程的训练/评估/预测流程中介绍的预测流程,即tools/infer.py是使用训练引擎分支的预测流程。保存的inference_model,可以通过`fluid.io.load_inference_model`接口,走训练引擎分支预测。本文档也同时介绍通过预测引擎的Python API进行预测,一般而言这种方式的速度优于前者。 在PaddlePaddle中预测引擎和训练引擎底层有着不同的优化方法,代码走不的分支,两者都可以进行预测。在入门教程的训练/评估/预测流程中介绍的预测流程,即tools/infer.py是使用训练引擎分支的预测流程。保存的inference_model,可以通过`fluid.io.load_inference_model`接口,走训练引擎分支预测。本文档也同时介绍通过预测引擎的Python API进行预测,一般而言这种方式的速度优于前者。
这篇教程介绍的Python API预测示例,除了可视化部分依赖PaddleDetection外,预处理、模型结构、执行流程均不依赖PaddleDetection。 这篇教程介绍的Python API预测示例,除了可视化部分依赖PaddleDetection外,预处理、模型结构、执行流程均不依赖PaddleDetection。
......
...@@ -55,7 +55,7 @@ def check_version(): ...@@ -55,7 +55,7 @@ def check_version():
"Please make sure the version is good with your code." \ "Please make sure the version is good with your code." \
try: try:
fluid.require_version('1.6.0') fluid.require_version('1.7.0')
except Exception as e: except Exception as e:
logger.error(err) logger.error(err)
sys.exit(1) sys.exit(1)
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册