From 153080b2236dd6a736974c7f20730ec87e874d3d Mon Sep 17 00:00:00 2001 From: Guanghua Yu <742925032@qq.com> Date: Fri, 25 Mar 2022 09:45:51 +0800 Subject: [PATCH] Update picodet benchmark (#5451) * update picodet benchmark --- configs/picodet/README.md | 35 ++++++++++++++++++++--------------- configs/picodet/README_cn.md | 36 +++++++++++++++++++++--------------- ppdet/engine/export_utils.py | 5 +++-- 3 files changed, 44 insertions(+), 32 deletions(-) diff --git a/configs/picodet/README.md b/configs/picodet/README.md index ca49a8ec3..a226cc9a9 100644 --- a/configs/picodet/README.md +++ b/configs/picodet/README.md @@ -6,7 +6,10 @@ English | [简体中文](README_cn.md) ## News -- Released a new series of PP-PicoDet models, it was used TAL/Task-aligned-Head and optimized PAN, which improved the accuracy and optimized CPU prediction speed. Moreover the training speed is greatly improved. **(2022.03.20)** +- Released a new series of PP-PicoDet models: **(2022.03.20)** + - (1) It was used TAL/Task-aligned-Head and optimized PAN, which greatly improved the accuracy; + - (2) Moreover optimized CPU prediction speed, and the training speed is greatly improved; + - (3) The export model includes post-processing, and the prediction directly outputs the result, without secondary development, and the migration cost is lower. ### Legacy Model @@ -28,24 +31,24 @@ We developed a series of lightweight models, named `PP-PicoDet`. Because of the ## Benchmark -| Model | Input size | mAPval
0.5:0.95 | mAPval
0.5 | Params
(M) | FLOPS
(G) | Latency[NCNN](#latency)
(ms) | Latency[Lite](#latency)
(ms) | Download | Config | +| Model | Input size | mAPval
0.5:0.95 | mAPval
0.5 | Params
(M) | FLOPS
(G) | Latency[CPU](#latency)
(ms) | Latency[Lite](#latency)
(ms) | Download | Config | | :-------- | :--------: | :---------------------: | :----------------: | :----------------: | :---------------: | :-----------------------------: | :-----------------------------: | :----------------------------------------: | :--------------------------------------- | -| PicoDet-XS | 320*320 | 23.5 | 36.1 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_xs_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_xs_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_xs_320_coco_lcnet.yml) | -| PicoDet-XS | 416*416 | 26.2 | 39.3 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_xs_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_xs_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_xs_416_coco_lcnet.yml) | -| PicoDet-S | 320*320 | 29.1 | 43.4 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_s_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_s_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_320_coco_lcnet.yml) | -| PicoDet-S | 416*416 | 32.5 | 47.6 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_s_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_s_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_416_coco_lcnet.yml) | -| PicoDet-M | 320*320 | 34.4 | 50.0 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_m_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_m_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_m_320_coco_lcnet.yml) | -| PicoDet-M | 416*416 | 37.5 | 53.4 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_m_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_m_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_m_416_coco_lcnet.yml) | -| PicoDet-L | 320*320 | 36.1 | 52.0 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_l_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_320_coco_lcnet.yml) | -| PicoDet-L | 416*416 | 39.4 | 55.7 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_l_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_416_coco_lcnet.yml) | -| PicoDet-L | 640*640 | 42.3 | 59.2 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_l_640_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_640_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_640_coco_lcnet.yml) | +| PicoDet-XS | 320*320 | 23.5 | 36.1 | 0.70 | 0.67 | 10.9ms | 7.81ms | [model](https://paddledet.bj.bcebos.com/models/picodet_xs_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_xs_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_xs_320_coco_lcnet.yml) | +| PicoDet-XS | 416*416 | 26.2 | 39.3 | 0.70 | 1.13 | 15.4ms | 12.38ms | [model](https://paddledet.bj.bcebos.com/models/picodet_xs_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_xs_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_xs_416_coco_lcnet.yml) | +| PicoDet-S | 320*320 | 29.1 | 43.4 | 1.18 | 0.97 | 12.6ms | 9.56ms | [model](https://paddledet.bj.bcebos.com/models/picodet_s_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_s_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_320_coco_lcnet.yml) | +| PicoDet-S | 416*416 | 32.5 | 47.6 | 1.18 | 1.65 | 17.2ms | 15.20 | [model](https://paddledet.bj.bcebos.com/models/picodet_s_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_s_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_416_coco_lcnet.yml) | +| PicoDet-M | 320*320 | 34.4 | 50.0 | 3.46 | 2.57 | 14.5ms | 17.68ms | [model](https://paddledet.bj.bcebos.com/models/picodet_m_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_m_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_m_320_coco_lcnet.yml) | +| PicoDet-M | 416*416 | 37.5 | 53.4 | 3.46 | 4.34 | 19.5ms | 28.39ms | [model](https://paddledet.bj.bcebos.com/models/picodet_m_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_m_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_m_416_coco_lcnet.yml) | +| PicoDet-L | 320*320 | 36.1 | 52.0 | 5.80 | 4.20 | 18.3ms | 25.21ms | [model](https://paddledet.bj.bcebos.com/models/picodet_l_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_320_coco_lcnet.yml) | +| PicoDet-L | 416*416 | 39.4 | 55.7 | 5.80 | 7.10 | 22.1ms | 42.23ms | [model](https://paddledet.bj.bcebos.com/models/picodet_l_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_416_coco_lcnet.yml) | +| PicoDet-L | 640*640 | 42.3 | 59.2 | 5.80 | 16.81 | 43.1ms | 108.1ms | [model](https://paddledet.bj.bcebos.com/models/picodet_l_640_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_640_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_640_coco_lcnet.yml) |
Table Notes: -- Latency: All our models test on `Qualcomm Snapdragon 865(4xA77+4xA55)` with 4 threads by arm8 and with FP16. In the above table, test latency on [NCNN](https://github.com/Tencent/ncnn) and `Lite`->[Paddle-Lite](https://github.com/PaddlePaddle/Paddle-Lite). And testing latency with code: [MobileDetBenchmark](https://github.com/JiweiMaster/MobileDetBenchmark). -- PicoDet is trained on COCO train2017 dataset and evaluated on COCO val2017. -- PicoDet used 4 or 8 GPUs for training and all checkpoints are trained with default settings and hyperparameters. +- Latency: All our models test on `Intel-Xeon-Gold-6148` CPU with MKLDNN by 10 threads and `Qualcomm Snapdragon 865(4xA77+4xA55)` with 4 threads by arm8 and with FP16. In the above table, test CPU latency on Paddle-Inference and testing Mobile latency with `Lite`->[Paddle-Lite](https://github.com/PaddlePaddle/Paddle-Lite). +- PicoDet is trained on COCO train2017 dataset and evaluated on COCO val2017. And PicoDet used 4 GPUs for training and all checkpoints are trained with default settings and hyperparameters. +- Benchmark test: When testing the speed benchmark, the post-processing is not included in the exported model, you need to set `-o export.benchmark=True` or manually modify [runtime.yml](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/configs/runtime.yml#L12).
@@ -65,6 +68,7 @@ We developed a series of lightweight models, named `PP-PicoDet`. Because of the | YOLOv5n | 640*640 | 28.4 | 46.0 | 1.9 | 4.5 | 40.35 | | YOLOv5s | 640*640 | 37.2 | 56.0 | 7.2 | 16.5 | 78.05 | +- Testing Mobile latency with code: [MobileDetBenchmark](https://github.com/JiweiMaster/MobileDetBenchmark). ## Quick Start @@ -93,12 +97,13 @@ We developed a series of lightweight models, named `PP-PicoDet`. Because of the export CUDA_VISIBLE_DEVICES=0 python tools/train.py -c configs/picodet/picodet_s_320_coco_lcnet.yml --eval ``` +If the GPU is out of memory during training, reduce the batch_size in TrainReader, and reduce the base_lr in LearningRate proportionally. - Training model on multi-GPU: ```shell -# training on single-GPU +# training on multi-GPU export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m paddle.distributed.launch --gpus 0,1,2,3,4,5,6,7 tools/train.py -c configs/picodet/picodet_s_320_coco_lcnet.yml --eval ``` diff --git a/configs/picodet/README_cn.md b/configs/picodet/README_cn.md index 9b7d880a5..7131200a2 100644 --- a/configs/picodet/README_cn.md +++ b/configs/picodet/README_cn.md @@ -6,7 +6,10 @@ ## 最新动态 -- 发布全新系列PP-PicoDet模型，引入TAL及Task-aligned Head，优化PAN等结构，精度大幅提升，优化CPU端预测速度，同时训练速度大幅提升。**（2022.03.20）** +- 发布全新系列PP-PicoDet模型：**（2022.03.20）** + - (1)引入TAL及Task-aligned Head，优化PAN等结构，精度大幅提升； + - (2)优化CPU端预测速度，同时训练速度大幅提升； + - (3)导出模型将后处理包含在网络中，预测直接输出box结果，无需二次开发，迁移成本更低。 ## 历史版本模型 @@ -32,23 +35,23 @@ PP-PicoDet模型有如下特点： | 模型 | 输入尺寸 | mAPval
0.5:0.95 | mAPval
0.5 | 参数量
(M) | FLOPS
(G) | 预测时延[NCNN](#latency)
(ms) | 预测时延[Lite](#latency)
(ms) | 下载 | 配置文件 | | :-------- | :--------: | :---------------------: | :----------------: | :----------------: | :---------------: | :-----------------------------: | :-----------------------------: | :----------------------------------------: | :--------------------------------------- | -| PicoDet-XS | 320*320 | 23.5 | 36.1 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_xs_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_xs_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_xs_320_coco_lcnet.yml) | -| PicoDet-XS | 416*416 | 26.2 | 39.3 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_xs_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_xs_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_xs_416_coco_lcnet.yml) | -| PicoDet-S | 320*320 | 29.1 | 43.4 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_s_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_s_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_320_coco_lcnet.yml) | -| PicoDet-S | 416*416 | 32.5 | 47.6 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_s_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_s_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_416_coco_lcnet.yml) | -| PicoDet-M | 320*320 | 34.4 | 50.0 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_m_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_m_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_m_320_coco_lcnet.yml) | -| PicoDet-M | 416*416 | 37.5 | 53.4 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_m_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_m_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_m_416_coco_lcnet.yml) | -| PicoDet-L | 320*320 | 36.1 | 52.0 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_l_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_320_coco_lcnet.yml) | -| PicoDet-L | 416*416 | 39.4 | 55.7 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_l_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_416_coco_lcnet.yml) | -| PicoDet-L | 640*640 | 42.3 | 59.2 | - | - | - | - | [model](https://paddledet.bj.bcebos.com/models/picodet_l_640_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_640_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_640_coco_lcnet.yml) | +| PicoDet-XS | 320*320 | 23.5 | 36.1 | 0.70 | 0.67 | 10.9ms | 7.81ms | [model](https://paddledet.bj.bcebos.com/models/picodet_xs_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_xs_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_xs_320_coco_lcnet.yml) | +| PicoDet-XS | 416*416 | 26.2 | 39.3 | 0.70 | 1.13 | 15.4ms | 12.38ms | [model](https://paddledet.bj.bcebos.com/models/picodet_xs_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_xs_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_xs_416_coco_lcnet.yml) | +| PicoDet-S | 320*320 | 29.1 | 43.4 | 1.18 | 0.97 | 12.6ms | 9.56ms | [model](https://paddledet.bj.bcebos.com/models/picodet_s_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_s_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_320_coco_lcnet.yml) | +| PicoDet-S | 416*416 | 32.5 | 47.6 | 1.18 | 1.65 | 17.2ms | 15.20 | [model](https://paddledet.bj.bcebos.com/models/picodet_s_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_s_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_s_416_coco_lcnet.yml) | +| PicoDet-M | 320*320 | 34.4 | 50.0 | 3.46 | 2.57 | 14.5ms | 17.68ms | [model](https://paddledet.bj.bcebos.com/models/picodet_m_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_m_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_m_320_coco_lcnet.yml) | +| PicoDet-M | 416*416 | 37.5 | 53.4 | 3.46 | 4.34 | 19.5ms | 28.39ms | [model](https://paddledet.bj.bcebos.com/models/picodet_m_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_m_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_m_416_coco_lcnet.yml) | +| PicoDet-L | 320*320 | 36.1 | 52.0 | 5.80 | 4.20 | 18.3ms | 25.21ms | [model](https://paddledet.bj.bcebos.com/models/picodet_l_320_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_320_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_320_coco_lcnet.yml) | +| PicoDet-L | 416*416 | 39.4 | 55.7 | 5.80 | 7.10 | 22.1ms | 42.23ms | [model](https://paddledet.bj.bcebos.com/models/picodet_l_416_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_416_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_416_coco_lcnet.yml) | +| PicoDet-L | 640*640 | 42.3 | 59.2 | 5.80 | 16.81 | 43.1ms | 108.1ms | [model](https://paddledet.bj.bcebos.com/models/picodet_l_640_coco_lcnet.pdparams) | [log](https://paddledet.bj.bcebos.com/logs/train_picodet_l_640_coco_lcnet.log) | [config](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/configs/picodet/picodet_l_640_coco_lcnet.yml) |
注意事项: -- 时延测试： 我们所有的模型都在`骁龙865(4xA77+4xA55)` 上测试(4线程，FP16预测)。上面表格中标有`NCNN`的是使用[NCNN](https://github.com/Tencent/ncnn)库测试，标有`Lite`的是使用[Paddle Lite](https://github.com/PaddlePaddle/Paddle-Lite)进行测试。 测试的benchmark脚本来自: [MobileDetBenchmark](https://github.com/JiweiMaster/MobileDetBenchmark)。 -- PicoDet在COCO train2017上训练，并且在COCO val2017上进行验证。 -- PicoDet使用4卡GPU训练(PicoDet-L-640使用8卡训练)，并且所有的模型都是通过发布的默认配置训练得到。 +- 时延测试： 我们所有的模型都在英特尔至强6148的CPU（MKLDNN 10线程）和`骁龙865(4xA77+4xA55)`的ARM CPU上测试(4线程，FP16预测)。上面表格中标有`CPU`的是使用Paddle Inference库测试，标有`Lite`的是使用[Paddle Lite](https://github.com/PaddlePaddle/Paddle-Lite)进行测试。 +- PicoDet在COCO train2017上训练，并且在COCO val2017上进行验证。使用4卡GPU训练，并且上表所有的预训练模型都是通过发布的默认配置训练得到。 +- Benchmark测试：测试速度benchmark性能时，导出模型后处理不包含在网络中，需要设置`-o export.benchmark=True` 或手动修改[runtime.yml](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/configs/runtime.yml#L12)。
@@ -68,13 +71,14 @@ PP-PicoDet模型有如下特点： | YOLOv5n | 640*640 | 28.4 | 46.0 | 1.9 | 4.5 | 40.35 | | YOLOv5s | 640*640 | 37.2 | 56.0 | 7.2 | 16.5 | 78.05 | +- ARM测试的benchmark脚本来自: [MobileDetBenchmark](https://github.com/JiweiMaster/MobileDetBenchmark)。 ## 快速开始
依赖包: -- PaddlePaddle >= 2.2.1 +- PaddlePaddle == 2.2.2
@@ -97,11 +101,13 @@ export CUDA_VISIBLE_DEVICES=0 python tools/train.py -c configs/picodet/picodet_s_320_coco_lcnet.yml --eval ``` +如果训练时显存out memory，将TrainReader中batch_size调小，同时LearningRate中base_lr等比例减小。 + - 多卡GPU上训练: ```shell -# training on single-GPU +# training on multi-GPU export CUDA_VISIBLE_DEVICES=0,1,2,3,4,5,6,7 python -m paddle.distributed.launch --gpus 0,1,2,3,4,5,6,7 tools/train.py -c configs/picodet/picodet_s_320_coco_lcnet.yml --eval ``` diff --git a/ppdet/engine/export_utils.py b/ppdet/engine/export_utils.py index 8f55f0293..bddb4af07 100644 --- a/ppdet/engine/export_utils.py +++ b/ppdet/engine/export_utils.py @@ -172,8 +172,9 @@ def _dump_infer_config(config, path, image_shape, model): reader_cfg, dataset_cfg, config['metric'], label_arch, image_shape[1:]) if infer_arch == 'PicoDet': - if hasattr(config, 'export') and config['export'].get('post_process', - False): + if hasattr(config, 'export') and config['export'].get( + 'post_process', + False) and not config['export'].get('benchmark', False): infer_cfg['arch'] = 'GFL' head_name = 'PicoHeadV2' if config['PicoHeadV2'] else 'PicoHead' infer_cfg['NMS'] = config[head_name]['nms'] -- GitLab