mirrors/PaddleClas
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/PaddleClas.git synced 2025-06-03 21:55:06 +08:00
Code Issues Projects Releases Wiki Activity

Update feature_extraction.md

Browse Source
This commit is contained in:
Bin Lu 2021-10-29 17:47:02 +08:00 committed by GitHub
parent bf92ff47de
commit adb0a78c4f
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
docs/zh_CN_tmp/image_recognition_pipeline/feature_extraction.md
View File

@ -78,7 +78,7 @@ Loss:
# 4.训练、评估、推理
下面以`ppcls/configs/Products/ResNet50_vd_SOP.yaml`为例，介绍模型的训练、评估、推理过程
## 4.1 数据准备
首先，下载SOP数据集， 数据链接
首先，下载SOP数据集， 数据链接： https://cvgl.stanford.edu/projects/lifted_struct/
## 4.2 训练
- 单机单卡训练
@ -96,7 +96,7 @@ python -m paddle.distributed.launch
## 4.3 评估
- 单卡评估
```
python tools/eval.py -c ppcls/configs/ResNet50_vd_SOP.yaml -o Global.pretrained_model = "output/ReModel/best_model"
python tools/eval.py -c ppcls/configs/ResNet50_vd_SOP.yaml -o Global.pretrained_model="output/ReModel/best_model"
```
- 多卡评估
```
@ -109,9 +109,9 @@ python -m paddle.distributed.launch
推理过程包括两个步骤： 1） 导出推理模型； 2） 获取特征向量
### 4.4.1 导出推理模型
```
python tools/export_model -c ppcls/configs/ResNet50_vd_SOP.yaml -o Global.pretrained_model = "output/ReModel/best_model"
python tools/export_model -c ppcls/configs/ResNet50_vd_SOP.yaml -o Global.pretrained_model="output/ReModel/best_model"
```
生成的推理模型位于inference目录，名字为inference.pd*
生成的推理模型位于`inference`目录，名字为`inference.pd*`
### 4.4.2 获取特征向量
```