花朵分类：从 PaddleLabel 到 PaddleClas
PaddleLabel 标注数据+PaddleClas 训练预测=快速完成一次花朵分类的任务
1. 数据准备
- 首先使用
PaddleLabel对自制的花朵数据集进行标注，其次使用Split Dataset功能分割数据集，最后导出数据集 - 从
PaddleLabel导出后的内容全部放到自己的建立的文件夹下，例如flower_clas_dataset，其目录结构如下：
├── flower_clas_dataset
│ ├── image
│ │ ├── flower1.jpg
│ │ ├── flower2.jpg
│ │ ├── ...
│ ├── labels.txt
│ ├── test_list.txt
│ ├── train_list.txt
│ ├── val_list.txt
2. 训练
2.1 安装必备的库
2.1.1 安装 paddlepaddle
# 您的机器安装的是 CUDA9 或 CUDA10，请运行以下命令安装
# pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple
# 您的机器是CPU，请运行以下命令安装
pip install paddlepaddle
2.1.2 安装 paddleclas 以及依赖项
git clone https://gitee.com/paddlepaddle/PaddleClas.git -b release/2.2
cd PaddleClas
pip install -r requirements.txt
python setup.py install
2.2 准备自制的花朵分类数据集
cd ./PaddleClas/dataset/
mkdir flower_clas_dataset
cd ../../
cp -r ./flower_clas_dataset/* ./PaddleClas/dataset/flower_clas_dataset
2.3 修改配置文件
PaddleClas/ppcls/configs/quick_start/new_user/ShuffleNetV2_x0_25.yaml
# global configs
Global:
checkpoints: null
pretrained_model: null
output_dir: ./output/
device: cpu
save_interval: 20
eval_during_train: True
eval_interval: 10
epochs: 100
print_batch_step: 10
use_visualdl: True
# used for static mode and model export
image_shape: [3, 224, 224]
save_inference_dir: ./inference
# model architecture
Arch:
name: ShuffleNetV2_x0_25
class_num: 3
# loss function config for traing/eval process
Loss:
Train:
- CELoss:
weight: 1.0
Eval:
- CELoss:
weight: 1.0
Optimizer:
name: Momentum
momentum: 0.9
lr:
name: Cosine
learning_rate: 0.0125
warmup_epoch: 5
regularizer:
name: 'L2'
coeff: 0.00001
# data loader for train and eval
DataLoader:
Train:
dataset:
name: ImageNetDataset
image_root: ./dataset/
cls_label_path: ./dataset/train_list.txt
transform_ops:
- DecodeImage:
to_rgb: True
channel_first: False
- RandCropImage:
size: 224
- RandFlipImage:
flip_code: 1
- NormalizeImage:
scale: 1.0/255.0
mean: [0.485, 0.456, 0.406]
std: [0.229, 0.224, 0.225]
order: ''
sampler:
name: DistributedBatchSampler
batch_size: 16
drop_last: False
shuffle: True
loader:
num_workers: 0
use_shared_memory: True
Eval:
dataset:
name: ImageNetDataset
image_root: ./dataset/
cls_label_path: ./dataset/val_list.txt
transform_ops:
- DecodeImage:
to_rgb: True
channel_first: False
- ResizeImage:
resize_short: 256
- CropImage:
size: 224
- NormalizeImage:
scale: 1.0/255.0
mean: [0.485, 0.456, 0.406]
std: [0.229, 0.224, 0.225]
order: ''
sampler:
name: DistributedBatchSampler
batch_size: 32
drop_last: False
shuffle: False
loader:
num_workers: 0
use_shared_memory: True
Infer:
infer_imgs: dataset/predict_demo.jpg
batch_size: 10
transforms:
- DecodeImage:
to_rgb: True
channel_first: False
- ResizeImage:
resize_short: 256
- CropImage:
size: 224
- NormalizeImage:
scale: 1.0/255.0
mean: [0.485, 0.456, 0.406]
std: [0.229, 0.224, 0.225]
order: ''
- ToCHWImage:
PostProcess:
name: Topk
topk: 3
Metric:
Train:
- TopkAcc:
topk: [1, 3]
Eval:
- TopkAcc:
topk: [1, 3]
2.4 添加类别映射文件
PaddleClas/ppcls/configs/quick_start/new_user/label.txt
sunflower
rose
dandelion
2.5 开始训练
export CUDA_VISIBLE_DEVICES=0
# 开始训练
python PaddleClas/tools/train.py -c ./PaddleClas/ppcls/configs/quick_start/new_user/ShuffleNetV2_x0_25.yaml
3. 模型评估
3.1 评估
python PaddleClas/tools/eval.py -c ./PaddleClas/ppcls/configs/quick_start/new_user/ShuffleNetV2_x0_25.yaml
3.2 预测
python3 PaddleClas/tools/infer.py \
-c ./PaddleClas/ppcls/configs/quick_start/new_user/ShuffleNetV2_x0_25.yaml \
-o Infer.infer_imgs=dataset/predict_demo.jpg \
-o Global.pretrained_model=output/ShuffleNetV2_x0_25/latest
预测的样例图片是：
预测的结果是：
{‘class_ids’: [0, 1, 2], ‘scores’: [0.89812, 0.09476, 0.00712], ‘file_name’: ‘dataset/predict_demo.jpg’, ‘label_names’: []} 也就是说 0 的概率最大，为 0.89812，0 对应的结果是向日葵，也就是说结果是向日葵，预测无误。