From 50e3bacd32e903fd2485b5bb49d8fa6dff276957 Mon Sep 17 00:00:00 2001 From: LDOUBLEV Date: Fri, 16 Sep 2022 12:05:42 +0800 Subject: [PATCH] fix slim doc --- deploy/slim/quantization/README.md | 16 +++------------- deploy/slim/quantization/README_en.md | 17 +++-------------- 2 files changed, 6 insertions(+), 27 deletions(-) diff --git a/deploy/slim/quantization/README.md b/deploy/slim/quantization/README.md index 4c1d784b..7f1ff7ae 100644 --- a/deploy/slim/quantization/README.md +++ b/deploy/slim/quantization/README.md @@ -22,7 +22,7 @@ ### 1. 安装PaddleSlim ```bash -pip3 install paddleslim==2.2.2 +pip3 install paddleslim==2.3.2 ``` ### 2. 准备训练好的模型 @@ -33,17 +33,7 @@ PaddleOCR提供了一系列训练好的[模型](../../../doc/doc_ch/models_list. 量化训练包括离线量化训练和在线量化训练，在线量化训练效果更好，需加载预训练模型，在定义好量化策略后即可对模型进行量化。 -量化训练的代码位于slim/quantization/quant.py 中，比如训练检测模型，训练指令如下： -```bash -python deploy/slim/quantization/quant.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.pretrained_model='your trained model' Global.save_model_dir=./output/quant_model - -# 比如下载提供的训练模型 -wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar -tar -xf ch_ppocr_mobile_v2.0_det_train.tar -python deploy/slim/quantization/quant.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.pretrained_model=./ch_ppocr_mobile_v2.0_det_train/best_accuracy Global.save_model_dir=./output/quant_model -``` - -模型蒸馏和模型量化可以同时使用，以PPOCRv3检测模型为例： +量化训练的代码位于slim/quantization/quant.py 中，比如训练检测模型，以PPOCRv3检测模型为例，训练指令如下： ``` # 下载检测预训练模型： wget https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_distill_train.tar @@ -58,7 +48,7 @@ python deploy/slim/quantization/quant.py -c configs/det/ch_PP-OCRv3/ch_PP-OCRv3_ 在得到量化训练保存的模型后，我们可以将其导出为inference_model，用于预测部署： ```bash -python deploy/slim/quantization/export_model.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.checkpoints=output/quant_model/best_accuracy Global.save_inference_dir=./output/quant_inference_model +python deploy/slim/quantization/export_model.py -c configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml -o Global.checkpoints=output/quant_model/best_accuracy Global.save_inference_dir=./output/quant_inference_model ``` ### 5. 量化模型部署 diff --git a/deploy/slim/quantization/README_en.md b/deploy/slim/quantization/README_en.md index c6796ae9..f82c3d84 100644 --- a/deploy/slim/quantization/README_en.md +++ b/deploy/slim/quantization/README_en.md @@ -25,7 +25,7 @@ After training, if you want to further compress the model size and accelerate th ### 1. Install PaddleSlim ```bash -pip3 install paddleslim==2.2.2 +pip3 install paddleslim==2.3.2 ``` @@ -39,18 +39,7 @@ Quantization training includes offline quantization training and online quantiza Online quantization training is more effective. It is necessary to load the pre-trained model. After the quantization strategy is defined, the model can be quantified. -The code for quantization training is located in `slim/quantization/quant.py`. For example, to train a detection model, the training instructions are as follows: -```bash -python deploy/slim/quantization/quant.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.pretrained_model='your trained model' Global.save_model_dir=./output/quant_model - -# download provided model -wget https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_det_train.tar -tar -xf ch_ppocr_mobile_v2.0_det_train.tar -python deploy/slim/quantization/quant.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.pretrained_model=./ch_ppocr_mobile_v2.0_det_train/best_accuracy Global.save_model_dir=./output/quant_model -``` - - -Model distillation and model quantization can be used at the same time, taking the PPOCRv3 detection model as an example: +The code for quantization training is located in `slim/quantization/quant.py`. For example, the training instructions of slim PPOCRv3 detection model are as follows: ``` # download provided model wget https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_distill_train.tar @@ -66,7 +55,7 @@ If you want to quantify the text recognition model, you can modify the configura Once we got the model after pruning and fine-tuning, we can export it as an inference model for the deployment of predictive tasks: ```bash -python deploy/slim/quantization/export_model.py -c configs/det/ch_ppocr_v2.0/ch_det_mv3_db_v2.0.yml -o Global.checkpoints=output/quant_model/best_accuracy Global.save_inference_dir=./output/quant_inference_model +python deploy/slim/quantization/export_model.py -c configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml -o Global.checkpoints=output/quant_model/best_accuracy Global.save_inference_dir=./output/quant_inference_model ``` ### 5. Deploy -- GitLab