Skip to content

P
PaddleDetection

s920243400 / PaddleDetection
与 Fork 源项目一致

Fork自 PaddlePaddle / PaddleDetection

通知 2
Star 0
Fork 0
P
PaddleDetection

体验新版 GitCode，发现更多精彩内容 >>

未验证 提交 91700aaa 编写于 作者: W wangguanzhong 提交者： GitHub
浏览文件
操作

polish doc (#286)

上级 441b3623
隐藏空白更改
内联 并排
Showing with 14 addition and 17 deletion
docs/featured_model/FACE_DETECTION.md
浏览文件 @ 91700aaa
......@@ -157,7 +157,7 @@ cd dataset/wider_face && ./download.sh
export CUDA_VISIBLE_DEVICES=0
export PYTHONPATH=$PYTHONPATH:.
python -u tools/face_eval.py -c configs/face_detection/blazeface.yml \
-o weights=output/blazeface/model_final/ \
-o weights=output/blazeface/model_final \
--eval_mode=widerface
```
评估完成后，将在`output/pred`中生成txt格式的测试结果。
......@@ -222,7 +222,7 @@ EvalReader:
评估并生成结果文件：
```
python -u tools/face_eval.py -c configs/face_detection/blazeface.yml \
-o weights=output/blazeface/model_final/ \
-o weights=output/blazeface/model_final \
--eval_mode=fddb
```
评估完成后，将在`output/pred/pred_fddb_res.txt`中生成txt格式的测试结果。
......
docs/featured_model/FACE_DETECTION_en.md
浏览文件 @ 91700aaa
......@@ -172,7 +172,7 @@ Default is `False`, it will select `single-scale` evaluation.
export CUDA_VISIBLE_DEVICES=0
export PYTHONPATH=$PYTHONPATH:.
python -u tools/face_eval.py -c configs/face_detection/blazeface.yml \
-o weights=output/blazeface/model_final/ \
-o weights=output/blazeface/model_final \
--eval_mode=widerface
```
After the evaluation is completed, the test result in txt format will be generated in `output/pred`.
......@@ -239,7 +239,7 @@ EvalReader:
Evaluate and generate results files:
```
python -u tools/face_eval.py -c configs/face_detection/blazeface.yml \
-o weights=output/blazeface/model_final/ \
-o weights=output/blazeface/model_final \
--eval_mode=fddb
```
After the evaluation is completed, the test result in txt format will be generated in `output/pred/pred_fddb_res.txt`.
......
docs/tutorials/GETTING_STARTED.md
浏览文件 @ 91700aaa
......@@ -29,7 +29,7 @@ list below can be viewed by `--help`
| FLAG | script supported | description | default | remark |
| :----------------------: | :------------: | :---------------: | :--------------: | :-----------------: |
| -c | ALL | Select config file | None | **The whole description of configure can refer to [config_example](config_example)** |
| -c | ALL | Select config file | None | **The description of configure can refer to [CONFIG.md](../advanced_tutorials/CONFIG.md)** |
| -o | ALL | Set parameters in configure file | None | `-o` has higher priority to file configured by `-c`. Such as `-o use_gpu=False max_iter=10000` |
| -r/--resume_checkpoint | train | Checkpoint path for resuming training | None | `-r output/faster_rcnn_r50_1x/10000` |
| --eval | train | Whether to perform evaluation in training | False | |
......@@ -70,7 +70,7 @@ list below can be viewed by `--help`
```bash
export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
python -u tools/train.py -c configs/faster_rcnn_r50_1x.yml \
-o pretrain_weights=output/faster_rcnn_r50_1x/model_final/ \
-o pretrain_weights=output/faster_rcnn_r50_1x/model_final \
finetune_exclude_pretrained_params = ['cls_score','bbox_pred']
```
......@@ -120,7 +120,6 @@ Also, please note mixed precision training currently requires changing `norm_typ
export CUDA_VISIBLE_DEVICES=0
python -u tools/eval.py -c configs/faster_rcnn_r50_1x.yml \
-o weights=https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_r50_1x.tar \
-d dataset/coco
```
The path of model to be evaluted can be both local path and link in [MODEL_ZOO](../MODEL_ZOO_cn.md).
......
docs/tutorials/GETTING_STARTED_cn.md
浏览文件 @ 91700aaa
......@@ -26,7 +26,7 @@ python tools/infer.py -c configs/faster_rcnn_r50_1x.yml --infer_img=demo/0000005
| FLAG | 支持脚本 | 用途 | 默认值 | 备注 |
| :----------------------: | :------------: | :---------------: | :--------------: | :-----------------: |
| -c | ALL | 指定配置文件 | None | **完整配置说明请参考[配置案例](config_example)** |
| -c | ALL | 指定配置文件 | None | **配置模块说明请参考[配置模块](../advanced_tutorials/CONFIG_cn.md)** |
| -o | ALL | 设置配置文件里的参数内容 | None | 使用-o配置相较于-c选择的配置文件具有更高的优先级。例如：`-o use_gpu=False max_iter=10000` |
| -r/--resume_checkpoint | train | 从某一检查点恢复训练 | None | `-r output/faster_rcnn_r50_1x/10000` |
| --eval | train | 是否边训练边测试 | False | |
......@@ -50,7 +50,7 @@ python tools/infer.py -c configs/faster_rcnn_r50_1x.yml --infer_img=demo/0000005
```bash
export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
python -u tools/train.py -c configs/faster_rcnn_r50_1x.yml --eval -d dataset/coco
python -u tools/train.py -c configs/faster_rcnn_r50_1x.yml --eval
```
在训练中交替执行评估, 评估在每个snapshot\_iter时开始。每次评估后还会评出最佳mAP模型保存到`best_model`文件夹下。
......@@ -68,7 +68,7 @@ python tools/infer.py -c configs/faster_rcnn_r50_1x.yml --infer_img=demo/0000005
```bash
export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7
python -u tools/train.py -c configs/faster_rcnn_r50_1x.yml \
-o pretrain_weights=output/faster_rcnn_r50_1x/model_final/ \
-o pretrain_weights=output/faster_rcnn_r50_1x/model_final \
finetune_exclude_pretrained_params=['cls_score','bbox_pred']
```
......@@ -118,10 +118,9 @@ python -m paddle.distributed.launch --selected_gpus 0,1,2,3,4,5,6,7 tools/train.
export CUDA_VISIBLE_DEVICES=0
python -u tools/eval.py -c configs/faster_rcnn_r50_1x.yml \
-o weights=https://paddlemodels.bj.bcebos.com/object_detection/faster_rcnn_r50_1x.tar \
-d dataset/coco
```
评估模型可以为本地路径，例如`output/faster_rcnn_r50_1x/model_final/`, 也可以是[MODEL_ZOO](../MODEL_ZOO_cn.md)中给出的模型链接。
评估模型可以为本地路径，例如`output/faster_rcnn_r50_1x/model_final`, 也可以是[MODEL_ZOO](../MODEL_ZOO_cn.md)中给出的模型链接。
- 通过json文件评估
......@@ -180,6 +179,5 @@ batch size可以达到每GPU 4 (Tesla V100 16GB)。
**A:** 可在配置文件中设置 `sample_transform`。注意需要在配置文件中加入**完整预处理**
例如RCNN模型中`DecodeImage`, `NormalizeImage` and `Permute`。
**Q:** affine_channel和batch norm是什么关系?
**A:** 在RCNN系列模型加载预训练模型初始化，有时候会固定住batch norm的参数, 使用预训练模型中的全局均值和方式，并且batch norm的scale和bias参数不更新，已发布的大多ResNet系列的RCNN模型采用这种方式。这种情况下可以在config中设置norm_type为bn或affine_channel, freeze_norm为true (默认为true)，两种方式等价。affne_channel的计算方式为`scale * x + bias`。只不过设置affine_channel时，内部对batch norm的参数自动做了融合。如果训练使用的affine_channel，用保存的模型做初始化，训练其他任务时，即可使用affine_channel, 也可使用batch norm, 参数均可正确加载。
**Q:** affine_channel和batch norm是什么关系? </br>
**A:** 在RCNN系列模型加载预训练模型初始化，有时候会固定住batch norm的参数, 使用预训练模型中的全局均值和方式，并且batch norm的scale和bias参数不更新，已发布的大多ResNet系列的RCNN模型采用这种方式。这种情况下可以在config中设置norm_type为bn或affine_channel, freeze_norm为true (默认为true)，两种方式等价。affne_channel的计算方式为`scale * x + bias`。只不过设置affine_channel时，内部对batch norm的参数自动做了融合。如果训练使用的affine_channel，用保存的模型做初始化，训练其他任务时，既可使用affine_channel, 也可使用batch norm, 参数均可正确加载。
docs/tutorials/QUICK_STARTED.md
浏览文件 @ 91700aaa
......@@ -13,7 +13,7 @@ export CUDA_VISIBLE_DEVICES=0
## Data Preparation
Dataset refers to [Kaggle](https://www.kaggle.com/mbkinaci/fruit-images-for-object-detection), which contains 240 images in train dataset and 60 images in test dataset. Data categories are apple, orange and banana. Download [here](https://dataset.bj.bcebos.com/PaddleDetection_demo/fruit-detection.tar) and uncompress the dataset after download, script for data preparation is located at [download_fruit.py](https://github.com/PaddlePaddle/PaddleDetection/blob/master/dataset/fruit/download_fruit.py). Command is as follows:
Dataset refers to [Kaggle](https://www.kaggle.com/mbkinaci/fruit-images-for-object-detection), which contains 240 images in train dataset and 60 images in test dataset. Data categories are apple, orange and banana. Download [here](https://dataset.bj.bcebos.com/PaddleDetection_demo/fruit-detection.tar) and uncompress the dataset after download, script for data preparation is located at [download_fruit.py](../../dataset/fruit/download_fruit.py). Command is as follows:
```bash
python dataset/fruit/download_fruit.py
......
docs/tutorials/QUICK_STARTED_cn.md
浏览文件 @ 91700aaa
......@@ -13,7 +13,7 @@ export CUDA_VISIBLE_DEVICES=0
## 数据准备
数据集参考[Kaggle数据集](https://www.kaggle.com/mbkinaci/fruit-images-for-object-detection)，其中训练数据集240张图片，测试数据集60张图片，数据类别为3类：苹果，橘子，香蕉。[下载链接](https://dataset.bj.bcebos.com/PaddleDetection_demo/fruit-detection.tar)。数据下载后分别解压即可, 数据准备脚本位于[download_fruit.py](../dataset/fruit/download_fruit.py)。下载数据方式如下：
数据集参考[Kaggle数据集](https://www.kaggle.com/mbkinaci/fruit-images-for-object-detection)，其中训练数据集240张图片，测试数据集60张图片，数据类别为3类：苹果，橘子，香蕉。[下载链接](https://dataset.bj.bcebos.com/PaddleDetection_demo/fruit-detection.tar)。数据下载后分别解压即可, 数据准备脚本位于[download_fruit.py](../../dataset/fruit/download_fruit.py)。下载数据方式如下：
```bash
python dataset/fruit/download_fruit.py
......
Markdown is supported
0% 或 .
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 或