未验证 提交 1a229859 编写于 作者: L leiqing 提交者: GitHub

Update README.md

上级 88711866
...@@ -22,8 +22,8 @@ ...@@ -22,8 +22,8 @@
| 模型 | Base mAP | 离线量化mAP | ACT量化mAP | TRT-FP32 | TRT-FP16 | TRT-INT8 | 配置文件 | 量化模型 | | 模型 | Base mAP | 离线量化mAP | ACT量化mAP | TRT-FP32 | TRT-FP16 | TRT-INT8 | 配置文件 | 量化模型 |
| :-------- |:-------- |:--------: | :---------------------: | :----------------: | :----------------: | :---------------: | :----------------------: | :---------------------: | | :-------- |:-------- |:--------: | :---------------------: | :----------------: | :----------------: | :---------------: | :----------------------: | :---------------------: |
| PP-YOLOE-l | 50.9 | - | 50.6 | 11.2ms | 7.7ms | **6.7ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/tree/develop/demo/auto_compression/detection/configs/ppyoloe_l_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/ppyoloe_crn_l_300e_coco_quant.tar) | | PP-YOLOE-l | 50.9 | - | 50.6 | 11.2ms | 7.7ms | **6.7ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/ppyoloe_l_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/ppyoloe_crn_l_300e_coco_quant.tar) |
| PP-YOLOE-s | 43.1 | 41.2 | 42.6 | 6.51ms | 2.77ms | **2.12ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/tree/develop/demo/auto_compression/detection/configs/ppyoloe_s_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/ppyoloe_s_quant.tar) | | PP-YOLOE-s | 43.1 | 41.2 | 42.6 | 6.51ms | 2.77ms | **2.12ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/ppyoloe_s_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/ppyoloe_s_quant.tar) |
- mAP的指标均在COCO val2017数据集中评测得到,IoU=0.5:0.95。 - mAP的指标均在COCO val2017数据集中评测得到,IoU=0.5:0.95。
- PP-YOLOE-l模型在Tesla V100的GPU环境下测试,并且开启TensorRT,batch_size=1,包含NMS,测试脚本是[benchmark demo](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.4/deploy/python) - PP-YOLOE-l模型在Tesla V100的GPU环境下测试,并且开启TensorRT,batch_size=1,包含NMS,测试脚本是[benchmark demo](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.4/deploy/python)
...@@ -39,9 +39,9 @@ ...@@ -39,9 +39,9 @@
安装paddlepaddle: 安装paddlepaddle:
```shell ```shell
# CPU # CPU
pip install paddlepaddle pip install paddlepaddle==2.3.2
# GPU # GPU 以Ubuntu、CUDA 11.2为例
pip install paddlepaddle-gpu python -m pip install paddlepaddle-gpu==2.3.2.post112 -f https://www.paddlepaddle.org.cn/whl/linux/mkl/avx/stable.html
``` ```
安装paddleslim: 安装paddleslim:
......
Markdown is supported
0% .
You are about to add 0 people to the discussion. Proceed with caution.
先完成此消息的编辑!
想要评论请 注册