windows 运行PaddleClas

windows下 运行PaddleClas

版本：

python: 3.7.1
paddlepaddle: 1.8.4

(1) 下载PaddleClas源码

https://github.com/PaddlePaddle/PaddleClas

(2) 进入工程主目录

cd path_to_PaddleClas

(3) 下载并解压flowers102数据集

cd dataset/flowers102
wget https://www.robots.ox.ac.uk/~vgg/data/flowers/102/102flowers.tgz
wget https://www.robots.ox.ac.uk/~vgg/data/flowers/102/imagelabels.mat
wget https://www.robots.ox.ac.uk/~vgg/data/flowers/102/setid.mat
tar -xf 102flowers.tgz

(4) 数据集划分

python generate_flowers102_list.py jpg train > train_list.txt
python generate_flowers102_list.py jpg valid > val_list.txt
python generate_flowers102_list.py jpg test > extra_list.txt
cat train_list.txt extra_list.txt > train_extra_list.txt

(5)通过tools/download.py下载所需要的预训练模型。

python tools/download.py -a ResNet50_vd -p ./pretrained -d True
python tools/download.py -a ResNet50_vd_ssld -p ./pretrained -d True
python tools/download.py -a MobileNetV3_large_x1_0 -p ./pretrained -d True

(6)注释掉tools/train_multi_platform.py

if __name__ == '__main__':
   # paddle.enable_static()
    args = parse_args()
    main(args)

(7) 运行训练程序脚本

set CUDA_VISIBLE_DEVICES=0
# CPU训练
# 不使用预训练模型
python tools/train_multi_platform.py -c configs/quick_start/ResNet50_vd.yaml -o model_save_dir=output -o use_gpu=False

# 使用预训练模型
python tools/train_multi_platform.py -c configs/quick_start/ResNet50_vd_finetune.yaml -o model_save_dir=G:\CV\output -o use_gpu=False -o vdl_dir=scalar

运行效果图：