README_en.md 3.8 KB
Newer Older
Y
yukavio 已提交
1 2


L
LDOUBLEV 已提交
3
## Introduction
Y
yukavio 已提交
4

L
LDOUBLEV 已提交
5
Complicated models help to improve the performance of the model, but it also leads to some redundancy in the model. Model tailoring reduces this redundancy by removing the sub-models in the network model, so as to reduce model calculation complexity and improve model inference performance. .
Y
yukavio 已提交
6

L
LDOUBLEV 已提交
7
This tutorial will introduce how to use PaddleSlim to crop PaddleOCR model.
Y
yukavio 已提交
8

L
LDOUBLEV 已提交
9 10 11 12
It is recommended that you could understand following pages before reading this example:
1. [PaddleOCR training methods](../../../doc/doc_ch/quickstart.md)
2. [The demo of prune](https://paddlepaddle.github.io/PaddleSlim/tutorials/pruning_tutorial/)
3. [PaddleSlim prune API](https://paddlepaddle.github.io/PaddleSlim/api/prune_api/)
Y
yukavio 已提交
13

L
LDOUBLEV 已提交
14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29
## Quick start

Five steps for OCR model prune:
1. Install PaddleSlim
2. Prepare the trained model
3. Sensitivity analysis and training
4. Model tailoring training
5. Export model, predict deployment

### 1. Install PaddleSlim

```bash
git clone https://github.com/PaddlePaddle/PaddleSlim.git
cd Paddleslim
python setup.py install
```
Y
yukavio 已提交
30 31


L
LDOUBLEV 已提交
32 33 34
### 2. Download Pretrain Model
Model prune needs to load pre-trained models.
PaddleOCR also provides a series of models [../../../doc/doc_en/models_list_en.md]. Developers can choose their own models or use their own models according to their needs.
Y
yukavio 已提交
35 36


L
LDOUBLEV 已提交
37
### 3. Pruning sensitivity analysis
Y
yukavio 已提交
38 39 40 41 42

  After the pre-training model is loaded, sensitivity analysis is performed on each network layer of the model to understand the redundancy of each network layer, thereby determining the pruning ratio of each network layer. For specific details of sensitivity analysis, see:[Sensitivity analysis](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/docs/zh_cn/tutorials/image_classification_sensitivity_analysis_tutorial.md)

Enter the PaddleOCR root directory,perform sensitivity analysis on the model with the following command:

Y
yukavio 已提交
43
```bash
Y
yukavio 已提交
44 45 46

python deploy/slim/prune/sensitivity_anal.py -c configs/det/det_mv3_db.yml -o Global.pretrain_weights=./deploy/slim/prune/pretrain_models/det_mv3_db/best_accuracy Global.test_batch_size_per_card=1

Y
yukavio 已提交
47
```
Y
yukavio 已提交
48 49 50



L
LDOUBLEV 已提交
51
### 4. Model pruning and Fine-tune
Y
yukavio 已提交
52 53 54

  When pruning, the previous sensitivity analysis file would determines the pruning ratio of each network layer. In the specific implementation, in order to retain as many low-level features extracted from the image as possible, we skipped the 4 convolutional layers close to the input in the backbone. Similarly, in order to reduce the model performance loss caused by pruning, we selected some of the less redundant and more sensitive [network layer](https://github.com/PaddlePaddle/PaddleOCR/blob/develop/deploy/slim/prune/pruning_and_finetune.py#L41) through the sensitivity table obtained from the previous sensitivity analysis.And choose to skip these network layers in the subsequent pruning process. After pruning, the model need a finetune process to recover the performance and the training strategy of finetune is similar to the strategy of training original OCR detection model.

Y
yukavio 已提交
55
```bash
Y
yukavio 已提交
56 57 58

python deploy/slim/prune/pruning_and_finetune.py -c configs/det/det_mv3_db.yml -o Global.pretrain_weights=./deploy/slim/prune/pretrain_models/det_mv3_db/best_accuracy Global.test_batch_size_per_card=1

Y
yukavio 已提交
59
```
Y
yukavio 已提交
60 61


L
LDOUBLEV 已提交
62
### 5.  Export inference model and deploy it
Y
yukavio 已提交
63

L
LDOUBLEV 已提交
64
We can export the pruned model as inference_model for deployment:
Y
yukavio 已提交
65
```bash
Y
yukavio 已提交
66
python deploy/slim/prune/export_prune_model.py -c configs/det/det_mv3_db.yml -o Global.pretrain_weights=./output/det_db/best_accuracy Global.test_batch_size_per_card=1 Global.save_inference_dir=inference_model
Y
yukavio 已提交
67
```
L
LDOUBLEV 已提交
68 69 70 71 72

Reference for prediction and deployment of inference model:
1. [inference model python prediction](../../../doc/doc_en/inference_en.md)
2. [inference model C++ prediction](../../cpp_infer/readme_en.md)
3. [Deployment of inference model on mobile](../../lite/readme_en.md)