From 7ed2aa7db81dc9db77cb1d8056ca4598eaf8cc42 Mon Sep 17 00:00:00 2001 From: Guanghua Yu <742925032@qq.com> Date: Wed, 12 Oct 2022 18:40:38 +0800 Subject: [PATCH] Update README.md --- example/full_quantization/picodet/README.md | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-) diff --git a/example/full_quantization/picodet/README.md b/example/full_quantization/picodet/README.md index 49578015..1f934f43 100644 --- a/example/full_quantization/picodet/README.md +++ b/example/full_quantization/picodet/README.md @@ -23,7 +23,7 @@ | 模型 | 策略 | mAP | TRT-FP32 | TRT-FP16 | TRT-INT8 | 配置文件 | 模型 | | :-------- |:-------- |:--------: | :----------------: | :----------------: | :---------------: | :----------------------: | :---------------------: | | PicoDet-S-NPU | Baseline | 30.1 | - | - | - | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_416_coco_npu.yml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/picodet_s_416_coco_npu.tar) | -| PicoDet-S-NPU | 量化训练 | 29.7 | - | - | - | [config](https://github.com/PaddlePaddle/PaddleSlim/tree/develop/demo/full_quantization/picodet/configs/picodet_npu_with_postprocess.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/picodet_s_npu_quant.tar) | +| PicoDet-S-NPU | 量化训练 | 29.7 | - | - | - | [config](https://github.com/PaddlePaddle/PaddleSlim/tree/develop/example/full_quantization/picodet/configs/picodet_npu_with_postprocess.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/picodet_s_npu_quant.tar) | - mAP的指标均在COCO val2017数据集中评测得到，IoU=0.5:0.95。 @@ -114,12 +114,19 @@ python run.py --config_path=./configs/picodet_npu.yaml --save_dir='./output/' #### 3.5 测试模型精度 -使用eval.py脚本得到模型的mAP： +- 使用eval.py脚本得到模型的mAP： ``` export CUDA_VISIBLE_DEVICES=0 python eval.py --config_path=./configs/picodet_npu_with_postprocess.yaml ``` +- 使用ONNXRuntime测试模型mAP： +``` +python onnxruntime_eval.py \ + --reader_config=configs/picodet_reader.yml \ + --model_path=picodet_s_416_coco_npu.onnx +``` + **注意**： - 要测试的模型路径可以在配置文件中`model_dir`字段下进行修改。 -- GitLab