18 KiB
PULC Classification Model of Containing or Uncontaining AD Code
Catalogue
- 1. Introduction
- 2. Quick Start
- 3. Training, Evaluation and Inference
- 4. Model Compression
- 5. SHAS
- 6. Inference Deployment
1. Introduction
This case provides a way for users to quickly build a lightweight, high-precision and practical classification model of AD code(Here the definition of AD code includes: QR code, bar code, mini apps code) exists using PaddleClas PULC (Practical Ultra Lightweight image Classification). The model can be widely used in live broadcast scenarios, audit scenarios, massive data filtering scenarios, etc.
The following table lists the relevant indicators of the model. The first two lines means that using SwinTransformer_tiny and MobileNetV3_small_x0_35 as the backbone to training. The third to sixth lines means that the backbone is replaced by PPLCNet, additional use of EDA strategy and additional use of EDA strategy and SKL-UGI knowledge distillation strategy.
| Backbone | Accuracy（%） | Latency（ms） | Size（M） | Training Strategy |
|---|---|---|---|---|
| SwinTranformer_tiny | 95.06 | 95.30 | 111 | using ImageNet pretrained model |
| MobileNetV3_small_x0_35 | 86.67 | 2.85 | 2.6 | using ImageNet pretrained model |
| PPLCNet_x1_0 | 93.64 | 2.13 | 7.0 | using ImageNet pretrained model |
| PPLCNet_x1_0 | 94.44 | 2.13 | 7.0 | using SSLD pretrained model |
| PPLCNet_x1_0 | 94.62 | 2.13 | 7.0 | using SSLD pretrained model + EDA strategy |
| PPLCNet_x1_0 | 94.94 | 2.13 | 7.0 | using SSLD pretrained model + EDA strategy + SKL-UGI knowledge distillation strategy |
It can be seen that high accuracy can be getted when backbone is SwinTranformer_tiny, but the speed is slow. Replacing backbone with the lightweight model MobileNetV3_small_x0_35, the speed can be greatly improved, but the accuracy will be greatly reduced. Replacing backbone with faster backbone PPLCNet_x1_0, the accuracy is higher more 7 percentage points than MobileNetv3_small_x0_35. At the same time, the speed can be more than 20% faster. After additional using the SSLD pretrained model, the accuracy can be improved by about 0.8 percentage points without affecting the inference speed. Further, additional using the EDA strategy, the accuracy can be increased by 0.2 percentage points. Finally, after additional using the SKL-UGI knowledge distillation, the accuracy can be further improved by 0.3 percentage points. At this point, the accuracy is close to that of SwinTranformer_tiny, but the speed is more than 40 times faster. The training method and deployment instructions of PULC will be introduced in detail below.
Note：
- The Latency is tested on Intel(R) Xeon(R) Gold 6148 CPU @ 2.40GHz. The MKLDNN is enabled and the number of threads is 10.
- About PP-LCNet, please refer to PP-LCNet Introduction and PP-LCNet Paper.
2. Quick Start
2.1 PaddlePaddle Installation
- Run the following command to install if CUDA9 or CUDA10 is available.
python3 -m pip install paddlepaddle-gpu -i https://mirror.baidu.com/pypi/simple
- Run the following command to install if GPU device is unavailable.
python3 -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
Please refer to PaddlePaddle Installation for more information about installation, for examples other versions.
2.2 PaddleClas wheel Installation
The command of PaddleClas installation as bellow:
pip3 install paddleclas
2.3 Prediction
First, please click here to download and unzip to get the test demo images.
- Prediction with CLI
paddleclas --model_name=code_exists --infer_imgs=pulc_demo_imgs/code_exists/contains_code_demo.jpg
Results:
>>> result
class_ids: [1], scores: [0.9955421453341842], label_names: ['contains_code'], filename: pulc_demo_imgs/code_exists/contains_code_demo.jpg
Predict complete!
Note:If you want to test other images, only need to specify the --infer_imgs argument, and the directory containing images is also supported.
- Prediction in Python
import paddleclas
model = paddleclas.PaddleClas(model_name="code_exists")
result = model.predict(input_data="pulc_demo_imgs/code_exists/contains_code_demo.jpg")
print(next(result))
Note: The result returned by model.predict() is a generator, so you need to use the next() function to call it or for loop to loop it. And it will predict with batch_size size batch and return the prediction results when called. The default batch_size is 1, and you also specify the batch_size when instantiating, such as model = paddleclas.PaddleClas(model_name = "code_exists", batch_size=2). The result of demo above:
>>> result
[{'class_ids': [1], 'scores': [0.9955421453341842], 'label_names': ['contains_code'], 'filename': 'pulc_demo_imgs/code_exists/contains_code_demo.jpg'}]
3. Training, Evaluation and Inference
3.1 Installation
- Please refer to Installation to get the description about installation.
3.2 Dataset
3.2.1 Dataset Introduction
As there is currently no publicly available data set for QR codes/applets/barcodes, the first step is to collect data that may contain AD codes. The collection method can be write a simple script to crawl the data. After collecting the data, we need to carry out a simple annotation, that is, to mark whether the image contains AD codes.
3.2.2 Getting Dataset
The data used in this case can be getted by processing the open source data. The detailed processes are as follows:
Some image of the processed dataset is as follows:
Where, 0 indicates that the image has no AD code, and 1 indicates that the image has AD code.
And you can also download the data processed directly.
cd path_to_PaddleClas
Enter the dataset/ directory, download and unzip the dataset.
cd dataset
wget https://paddleclas.bj.bcebos.com/data/PULC/code_exists.tar
tar -xf code_exists.tar
cd ../
The datas under code_exists directory:
├── train
│ ├── 0
│ │ ├── 10021.jpg
│ │ ├── 10038.jpg
│ │ ├── ...
│ └── 1
│ ├── 10016.jpg
│ ├── 10028.jpg
│ ├── ...
├── train_list.txt
├── val
│ ├── 0
│ │ ├── 10077.jpg
│ │ ├── 10184.jpg
│ │ ├── ...
│ └── 1
│ ├── 10091.jpg
│ ├── 10098.jpg
│ ├── ...
└── val_list.txt
Where train/ and val/ are training set and validation set respectively. The train_list.txt and val_list.txt are label files of training data and validation data respectively.
Note：
-
About the contents format of
train_list.txtandval_list.txt, please refer to Description about Classification Dataset in PaddleClas. -
About the
train_list_for_distill.txt, please refer to Knowledge Distillation Label.
3.3 Training
The details of training config in ppcls/configs/PULC/code_exists/PPLCNet_x1_0.yaml. The command about training as follows:
export CUDA_VISIBLE_DEVICES=0,1,2,3
python3 -m paddle.distributed.launch \
--gpus="0,1,2,3" \
tools/train.py \
-c ./ppcls/configs/PULC/code_exists/PPLCNet_x1_0.yaml
The best metric of validation data is between 0.93 and 0.94. There would be fluctuations because the data size is small.
Note: Since a subset of the full data is used here, the indicator will be slightly lower, as shown below. The pre-training model and inference model provided in this case are obtained through full data training, which can be downloaded and used directly.
3.4 Evaluation
After training, you can use the following commands to evaluate the model.
python3 tools/eval.py \
-c ./ppcls/configs/PULC/code_exists/PPLCNet_x1_0.yaml \
-o Global.pretrained_model="output/PPLCNet_x1_0/best_model"
Among the above command, the argument -o Global.pretrained_model="output/PPLCNet_x1_0/best_model" specify the path of the best model weight file. You can specify other path if needed.
3.5 Inference
After training, you can use the model that trained to infer. Command is as follow:
python3 tools/infer.py \
-c ./ppcls/configs/PULC/code_exists/PPLCNet_x1_0.yaml \
-o Global.pretrained_model=output/PPLCNet_x1_0/best_model
The results:
[{'class_ids': [1], 'scores': [0.9999976], 'label_names': ['contains_code'], 'file_name': 'deploy/images/PULC/code_exists/contains_code_demo.jpg'}]
Note：
-
Among the above command, argument
-o Global.pretrained_model="output/PPLCNet_x1_0/best_model"specify the path of the best model weight file. You can specify other path if needed. -
The default test image is
deploy/images/PULC/code_exists/contains_code_demo.jpg. And you can test other image, only need to specify the argument-o Infer.infer_imgs=xxx.
4. Model Compression
4.1 SKL-UGI Knowledge Distillation
SKL-UGI is a simple but effective knowledge distillation algrithem proposed by PaddleClas.
4.1.1 Teacher Model Training
Training the teacher model with hyperparameters specified in ppcls/configs/PULC/code_exists/PPLCNet/PPLCNet_x1_0.yaml. The command is as follow:
export CUDA_VISIBLE_DEVICES=0,1,2,3
python3 -m paddle.distributed.launch \
--gpus="0,1,2,3" \
tools/train.py \
-c ./ppcls/configs/PULC/code_exists/PPLCNet_x1_0.yaml \
-o Arch.name=ResNet101_vd
The best metric of validation data is between 0.95 and 0.96. The best teacher model weight would be saved in file output/ResNet101_vd/best_model.pdparams.
4.1.2 Knowledge Distillation Training
The training strategy, specified in training config file ppcls/configs/PULC/code_exists/PPLCNet_x1_0_distillation.yaml, the teacher model is ResNet101_vd, the student model is PPLCNet_x1_0 and the additional unlabeled training data is validation data of ImageNet1k. The command is as follow:
export CUDA_VISIBLE_DEVICES=0,1,2,3
python3 -m paddle.distributed.launch \
--gpus="0,1,2,3" \
tools/train.py \
-c ./ppcls/configs/PULC/code_exists/PPLCNet_x1_0_distillation.yaml \
-o Arch.models.0.Teacher.pretrained=output/ResNet101_vd/best_model
The best metric is between 0.945 and 0.955. The best student model weight would be saved in file output/DistillationModel/best_model_student.pdparams.
5. Hyperparameters Searching
The hyperparameters used by 3.3 section and 4.1 section are according by Hyperparameters Searching in PaddleClas. If you want to get better results on your own dataset, you can refer to Hyperparameters Searching to get better hyperparameters.
Note: This section is optional. Because the search process will take a long time, you can selectively run according to your specific. If not replace the dataset, you can ignore this section.
6. Inference Deployment
6.1 Getting Paddle Inference Model
Paddle Inference is the original Inference Library of the PaddlePaddle, provides high-performance inference for server deployment. And compared with directly based on the pretrained model, Paddle Inference can use tools to accelerate prediction, so as to achieve better inference performance. Please refer to Paddle Inference for more information.
Paddle Inference need Paddle Inference Model to predict. Two process provided to get Paddle Inference Model. If want to use the provided by PaddleClas, you can download directly, click Downloading Inference Model.
6.1.1 Exporting Paddle Inference Model
The command about exporting Paddle Inference Model is as follow:
python3 tools/export_model.py \
-c ./ppcls/configs/PULC/code_exists/PPLCNet_x1_0.yaml \
-o Global.pretrained_model=output/DistillationModel/best_model_student \
-o Global.save_inference_dir=deploy/models/PPLCNet_x1_0_code_exists_infer
After running above command, the inference model files would be saved in deploy/models/PPLCNet_x1_0_code_exists_infer, as shown below:
├── PPLCNet_x1_0_code_exists_infer
│ ├── inference.pdiparams
│ ├── inference.pdiparams.info
│ └── inference.pdmodel
Note: The best model is from knowledge distillation training. If knowledge distillation training is not used, the best model would be saved in output/PPLCNet_x1_0/best_model.pdparams.
6.1.2 Downloading Inference Model
You can also download directly.
cd deploy/models
# download the inference model and decompression
wget https://paddleclas.bj.bcebos.com/models/PULC/code_exists_infer.tar && tar -xf code_exists_infer.tar
After decompression, the directory models should be shown below.
├── code_exists_infer
│ ├── inference.pdiparams
│ ├── inference.pdiparams.info
│ └── inference.pdmodel
6.2 Prediction with Python
6.2.1 Image Prediction
Return the directory deploy:
cd ../
Run the following command to classify whether there are AD codes in the image ./images/PULC/code_exists/contains_code_demo.jpg
# Use the following command to predict with GPU.
python3.7 python/predict_cls.py -c configs/PULC/code_exists/inference_code_exists.yaml
# Use the following command to predict with CPU.
python3.7 python/predict_cls.py -c configs/PULC/code_exists/inference_code_exists.yaml -o Global.use_gpu=False
The prediction results:
contains_code.jpg: class id(s): [1], score(s): [1.00], label_name(s): ['contains_code']
6.2.2 Images Prediction
If you want to predict images in directory, please specify the argument Global.infer_imgs as directory path by -o Global.infer_imgs. The command is as follow.
# Use the following command to predict with GPU. If want to replace with CPU, you can add argument -o Global.use_gpu=False
python3.7 python/predict_cls.py -c configs/PULC/code_exists/inference_code_exists.yaml -o Global.infer_imgs="./images/PULC/code_exists/"
All prediction results will be printed, as shown below.
no_code_demo.jpg: class id(s): [0], score(s): [1.00], label_name(s): ['no_code']
contains_code_demo.jpg: class id(s): [1], score(s): [1.00], label_name(s): ['contains_code']
Among the prediction results above,contains_code means that there are AD codes in the image,no-code means that there is no AD code in the image.
6.3 Deployment with C++
PaddleClas provides an example about how to deploy with C++. Please refer to Deployment with C++.
6.4 Deployment as Service
Paddle Serving is a flexible, high-performance carrier for machine learning models, and supports different protocol, such as RESTful, gRPC, bRPC and so on, which provides different deployment solutions for a variety of heterogeneous hardware and operating system environments. Please refer Paddle Serving for more information.
PaddleClas provides an example about how to deploy as service by Paddle Serving. Please refer to Paddle Serving Deployment.
6.5 Deployment on Mobile
Paddle-Lite is an open source deep learning framework that designed to make easy to perform inference on mobile, embeded, and IoT devices. Please refer to Paddle-Lite for more information.
PaddleClas provides an example of how to deploy on mobile by Paddle-Lite. Please refer to Paddle-Lite deployment.
6.6 Converting To ONNX and Deployment
Paddle2ONNX support convert Paddle Inference model to ONNX model. And you can deploy with ONNX model on different inference engine, such as TensorRT, OpenVINO, MNN/TNN, NCNN and so on. About Paddle2ONNX details, please refer to Paddle2ONNX.
PaddleClas provides an example of how to convert Paddle Inference model to ONNX model by paddle2onnx toolkit and predict by ONNX model. You can refer to paddle2onnx for deployment details.