更换文档检测模型
This commit is contained in:
186
paddle_detection/deploy/auto_compression/README.md
Normal file
186
paddle_detection/deploy/auto_compression/README.md
Normal file
@@ -0,0 +1,186 @@
|
||||
# 自动化压缩
|
||||
|
||||
目录：
|
||||
- [1.简介](#1简介)
|
||||
- [2.Benchmark](#2Benchmark)
|
||||
- [3.开始自动压缩](#自动压缩流程)
|
||||
- [3.1 环境准备](#31-准备环境)
|
||||
- [3.2 准备数据集](#32-准备数据集)
|
||||
- [3.3 准备预测模型](#33-准备预测模型)
|
||||
- [3.4 测试模型精度](#34-测试模型精度)
|
||||
- [3.5 自动压缩并产出模型](#35-自动压缩并产出模型)
|
||||
- [4.预测部署](#4预测部署)
|
||||
|
||||
## 1. 简介
|
||||
本示例使用PaddleDetection中Inference部署模型进行自动化压缩，使用的自动化压缩策略为量化蒸馏。
|
||||
|
||||
|
||||
## 2.Benchmark
|
||||
|
||||
### PP-YOLOE+
|
||||
|
||||
| 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+_s | 43.7 | - | 42.9 | - | - | - | [config](./configs/ppyoloe_plus_s_qat_dis.yaml) | [Quant Model](https://bj.bcebos.com/v1/paddledet/deploy/Inference/ppyoloe_plus_s_qat_dis.tar) |
|
||||
| PP-YOLOE+_m | 49.8 | - | 49.3 | - | - | - | [config](./configs/ppyoloe_plus_m_qat_dis.yaml) | [Quant Model](https://bj.bcebos.com/v1/paddledet/deploy/Inference/ppyoloe_plus_m_qat_dis.tar) |
|
||||
| PP-YOLOE+_l | 52.9 | - | 52.6 | - | - | - | [config](./configs/ppyoloe_plus_l_qat_dis.yaml) | [Quant Model](https://bj.bcebos.com/v1/paddledet/deploy/Inference/ppyoloe_plus_l_qat_dis.tar) |
|
||||
| PP-YOLOE+_x | 54.7 | - | 54.4 | - | - | - | [config](./configs/ppyoloe_plus_x_qat_dis.yaml) | [Quant Model](https://bj.bcebos.com/v1/paddledet/deploy/Inference/ppyoloe_plus_x_qat_dis.tar) |
|
||||
|
||||
- mAP的指标均在COCO val2017数据集中评测得到，IoU=0.5:0.95。
|
||||
|
||||
### YOLOv8
|
||||
|
||||
| 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 |
|
||||
| :-------- |:-------- |:--------: | :---------------------: | :----------------: | :----------------: | :---------------: | :----------------------: | :---------------------: |
|
||||
| YOLOv8-s | 44.9 | 43.9 | 44.3 | 9.27ms | 4.65ms | **3.78ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/yolov8_s_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/yolov8_s_500e_coco_trt_nms_quant.tar) |
|
||||
|
||||
**注意：**
|
||||
- 表格中YOLOv8模型均为带NMS的模型，可直接在TRT中部署，如果需要对齐测试标准，需要测试不带NMS的模型。
|
||||
- mAP的指标均在COCO val2017数据集中评测得到，IoU=0.5:0.95。
|
||||
- 表格中的性能在Tesla T4的GPU环境下测试，并且开启TensorRT，batch_size=1。
|
||||
|
||||
### PP-YOLOE
|
||||
|
||||
| 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/PaddleDetection/tree/develop/deploy/auto_compression/configs/ppyoloe_l_qat_dis.yaml) | [Quant Model](https://bj.bcebos.com/v1/paddle-slim-models/act/ppyoloe_crn_l_300e_coco_quant.tar) |
|
||||
| PP-YOLOE-SOD | 38.5 | - | 37.6 | - | - | - | [config](./configs/ppyoloe_crn_l_80e_sliced_visdrone_640_025_qat.yml) | [Quant Model](https://bj.bcebos.com/v1/paddle-slim-models/act/ppyoloe_sod_visdrone.tar) |
|
||||
|
||||
git
|
||||
- PP-YOLOE-l mAP的指标在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-SOD 的指标在VisDrone-DET数据集切图后的COCO格式[数据集](https://bj.bcebos.com/v1/paddledet/data/smalldet/visdrone_sliced.zip)中评测得到，IoU=0.5:0.95。定义文件[ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml](../../configs/smalldet/ppyoloe_crn_l_80e_sliced_visdrone_640_025.yml)
|
||||
|
||||
### PP-PicoDet
|
||||
|
||||
| 模型 | 策略 | mAP | FP32 | FP16 | 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/detection/configs/picodet_s_qat_dis.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。
|
||||
|
||||
### RT-DETR
|
||||
|
||||
| 模型 | Base mAP | ACT量化mAP | TRT-FP32 | TRT-FP16 | TRT-INT8 | 配置文件 | 量化模型 |
|
||||
| :---------------- | :------- | :--------: | :------: | :------: | :--------: | :----------------------------------------------------------: | :----------------------------------------------------------: |
|
||||
| RT-DETR-R50 | 53.1 | 53.0 | 32.05ms | 9.12ms | **6.96ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/rtdetr_r50vd_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/rtdetr_r50vd_6x_coco_quant.tar) |
|
||||
| RT-DETR-R101 | 54.3 | 54.1 | 54.13ms | 12.68ms | **9.20ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/rtdetr_r101vd_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/rtdetr_r101vd_6x_coco_quant.tar) |
|
||||
| RT-DETR-HGNetv2-L | 53.0 | 52.9 | 26.16ms | 8.54ms | **6.65ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/rtdetr_hgnetv2_l_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/rtdetr_hgnetv2_l_6x_coco_quant.tar) |
|
||||
| RT-DETR-HGNetv2-X | 54.8 | 54.6 | 49.22ms | 12.50ms | **9.24ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/rtdetr_hgnetv2_x_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/rtdetr_hgnetv2_x_6x_coco_quant.tar) |
|
||||
|
||||
- 上表测试环境：Tesla T4，TensorRT 8.6.0，CUDA 11.7，batch_size=1。
|
||||
|
||||
| 模型 | Base mAP | ACT量化mAP | TRT-FP32 | TRT-FP16 | TRT-INT8 | 配置文件 | 量化模型 |
|
||||
| :---------------- | :------- | :--------: | :------: | :------: | :--------: | :----------------------------------------------------------: | :----------------------------------------------------------: |
|
||||
| RT-DETR-R50 | 53.1 | 53.0 | 9.64ms | 5.00ms | **3.99ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/rtdetr_r50vd_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/rtdetr_r50vd_6x_coco_quant.tar) |
|
||||
| RT-DETR-R101 | 54.3 | 54.1 | 14.93ms | 7.15ms | **5.12ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/rtdetr_r101vd_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/rtdetr_r101vd_6x_coco_quant.tar) |
|
||||
| RT-DETR-HGNetv2-L | 53.0 | 52.9 | 8.17ms | 4.77ms | **4.00ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/rtdetr_hgnetv2_l_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/rtdetr_hgnetv2_l_6x_coco_quant.tar) |
|
||||
| RT-DETR-HGNetv2-X | 54.8 | 54.6 | 12.81ms | 6.97ms | **5.32ms** | [config](https://github.com/PaddlePaddle/PaddleSlim/blob/develop/example/auto_compression/detection/configs/rtdetr_hgnetv2_x_qat_dis.yaml) | [Model](https://bj.bcebos.com/v1/paddle-slim-models/act/rtdetr_hgnetv2_x_6x_coco_quant.tar) |
|
||||
|
||||
- 上表测试环境：A10，TensorRT 8.6.0，CUDA 11.6，batch_size=1。
|
||||
- mAP的指标均在COCO val2017数据集中评测得到，IoU=0.5:0.95。
|
||||
|
||||
## 3. 自动压缩流程
|
||||
|
||||
#### 3.1 准备环境
|
||||
- PaddlePaddle >= 2.4 （可从[Paddle官网](https://www.paddlepaddle.org.cn/install/quick?docurl=/documentation/docs/zh/install/pip/linux-pip.html)下载安装）
|
||||
- PaddleSlim >= 2.4.1
|
||||
- PaddleDet >= 2.5
|
||||
- opencv-python
|
||||
|
||||
安装paddlepaddle：
|
||||
```shell
|
||||
# CPU
|
||||
pip install paddlepaddle
|
||||
# GPU
|
||||
pip install paddlepaddle-gpu
|
||||
```
|
||||
|
||||
安装paddleslim：
|
||||
```shell
|
||||
pip install paddleslim
|
||||
```
|
||||
|
||||
安装paddledet：
|
||||
```shell
|
||||
pip install paddledet
|
||||
```
|
||||
|
||||
**注意：** YOLOv8模型的自动化压缩需要依赖安装最新[Develop Paddle](https://www.paddlepaddle.org.cn/install/quick?docurl=/documentation/docs/zh/develop/install/pip/linux-pip.html)和[Develop PaddleSlim](https://github.com/PaddlePaddle/PaddleSlim#%E5%AE%89%E8%A3%85)版本。
|
||||
|
||||
#### 3.2 准备数据集
|
||||
|
||||
本案例默认以COCO数据进行自动压缩实验，如果自定义COCO数据，或者其他格式数据，请参考[数据准备文档](https://github.com/PaddlePaddle/PaddleDetection/tree/develop/docs/tutorials/data/PrepareDataSet.md) 来准备数据。
|
||||
|
||||
如果数据集为非COCO格式数据，请修改[configs](./configs)中reader配置文件中的Dataset字段。
|
||||
|
||||
以PP-YOLOE模型为例，如果已经准备好数据集，请直接修改[./configs/yolo_reader.yml]中`EvalDataset`的`dataset_dir`字段为自己数据集路径即可。
|
||||
|
||||
#### 3.3 准备预测模型
|
||||
|
||||
预测模型的格式为：`model.pdmodel` 和 `model.pdiparams`两个，带`pdmodel`的是模型文件，带`pdiparams`后缀的是权重文件。
|
||||
|
||||
|
||||
根据[PaddleDetection文档](https://github.com/PaddlePaddle/PaddleDetection/blob/develop/docs/tutorials/GETTING_STARTED_cn.md#8-%E6%A8%A1%E5%9E%8B%E5%AF%BC%E5%87%BA) 导出Inference模型，具体可参考下方PP-YOLOE模型的导出示例：
|
||||
- 下载代码
|
||||
```
|
||||
git clone https://github.com/PaddlePaddle/PaddleDetection.git
|
||||
```
|
||||
- 导出预测模型
|
||||
|
||||
PPYOLOE-l模型，包含NMS：如快速体验，可直接下载[PP-YOLOE-l导出模型](https://bj.bcebos.com/v1/paddle-slim-models/act/ppyoloe_crn_l_300e_coco.tar)
|
||||
```shell
|
||||
python tools/export_model.py \
|
||||
-c configs/ppyoloe/ppyoloe_crn_l_300e_coco.yml \
|
||||
-o weights=https://paddledet.bj.bcebos.com/models/ppyoloe_crn_l_300e_coco.pdparams \
|
||||
trt=True \
|
||||
```
|
||||
|
||||
YOLOv8-s模型，包含NMS，具体可参考[YOLOv8模型文档](https://github.com/PaddlePaddle/PaddleYOLO/tree/release/2.5/configs/yolov8), 然后执行：
|
||||
```shell
|
||||
python tools/export_model.py \
|
||||
-c configs/yolov8/yolov8_s_500e_coco.yml \
|
||||
-o weights=https://paddledet.bj.bcebos.com/models/yolov8_s_500e_coco.pdparams \
|
||||
trt=True
|
||||
```
|
||||
|
||||
如快速体验，可直接下载[YOLOv8-s导出模型](https://bj.bcebos.com/v1/paddle-slim-models/act/yolov8_s_500e_coco_trt_nms.tar)
|
||||
|
||||
#### 3.4 自动压缩并产出模型
|
||||
|
||||
蒸馏量化自动压缩示例通过run.py脚本启动，会使用接口```paddleslim.auto_compression.AutoCompression```对模型进行自动压缩。配置config文件中模型路径、蒸馏、量化、和训练等部分的参数，配置完成后便可对模型进行量化和蒸馏。具体运行命令为：
|
||||
|
||||
- 单卡训练：
|
||||
```
|
||||
export CUDA_VISIBLE_DEVICES=0
|
||||
python run.py --config_path=./configs/ppyoloe_l_qat_dis.yaml --save_dir='./output/'
|
||||
```
|
||||
|
||||
- 多卡训练：
|
||||
```
|
||||
CUDA_VISIBLE_DEVICES=0,1,2,3 python -m paddle.distributed.launch --log_dir=log --gpus 0,1,2,3 run.py \
|
||||
--config_path=./configs/ppyoloe_l_qat_dis.yaml --save_dir='./output/'
|
||||
```
|
||||
|
||||
#### 3.5 测试模型精度
|
||||
|
||||
使用eval.py脚本得到模型的mAP：
|
||||
```
|
||||
export CUDA_VISIBLE_DEVICES=0
|
||||
python eval.py --config_path=./configs/ppyoloe_l_qat_dis.yaml
|
||||
```
|
||||
|
||||
使用paddle inference并使用trt int8得到模型的mAP:
|
||||
```
|
||||
export CUDA_VISIBLE_DEVICES=0
|
||||
python paddle_inference_eval.py --model_path ./output/ --reader_config configs/ppyoloe_reader.yml --precision int8 --use_trt=True
|
||||
```
|
||||
|
||||
**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。
|
||||
- --precision 默认为paddle，如果使用trt，需要设置--use_trt=True，同时--precision 可设置为fp32/fp16/int8
|
||||
|
||||
## 4.预测部署
|
||||
|
||||
- 可以参考[PaddleDetection部署教程](https://github.com/PaddlePaddle/PaddleDetection/tree/release/2.4/deploy)，GPU上量化模型开启TensorRT并设置trt_int8模式进行部署。
|
||||
@@ -0,0 +1,32 @@
|
||||
metric: COCO
|
||||
num_classes: 80
|
||||
|
||||
|
||||
# Datset configuration
|
||||
TrainDataset:
|
||||
!COCODataSet
|
||||
image_dir: train2017
|
||||
anno_path: annotations/instances_train2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
EvalDataset:
|
||||
!COCODataSet
|
||||
image_dir: val2017
|
||||
anno_path: annotations/instances_val2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
worker_num: 6
|
||||
eval_height: &eval_height 416
|
||||
eval_width: &eval_width 416
|
||||
eval_size: &eval_size [*eval_height, *eval_width]
|
||||
|
||||
EvalReader:
|
||||
sample_transforms:
|
||||
- Decode: {}
|
||||
- Resize: {interp: 2, target_size: *eval_size, keep_ratio: False}
|
||||
- NormalizeImage: {mean: [0, 0, 0], std: [1, 1, 1], is_scale: True}
|
||||
- Permute: {}
|
||||
batch_transforms:
|
||||
- PadBatch: {pad_to_stride: 32}
|
||||
batch_size: 8
|
||||
shuffle: false
|
||||
@@ -0,0 +1,34 @@
|
||||
Global:
|
||||
reader_config: ./configs/picodet_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./picodet_s_416_coco_npu/
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: l2
|
||||
|
||||
QuantAware:
|
||||
use_pact: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
weight_bits: 8
|
||||
activation_bits: 8
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 8000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00001
|
||||
T_max: 8000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
|
||||
|
||||
@@ -0,0 +1,34 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/ppyoloe_crn_l_80e_sliced_visdrone_640_025_reader.yml
|
||||
input_list: ['image', 'scale_factor']
|
||||
arch: YOLO
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ../../output_inference/ppyoloe_crn_l_80e_sliced_visdrone_640_025
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
onnx_format: True
|
||||
use_pact: False
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 8000
|
||||
eval_iter: 500
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 6000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
@@ -0,0 +1,25 @@
|
||||
metric: COCO
|
||||
num_classes: 10
|
||||
|
||||
# Datset configuration
|
||||
TrainDataset:
|
||||
!COCODataSet
|
||||
image_dir: train_images_640_025
|
||||
anno_path: train_640_025.json
|
||||
dataset_dir: dataset/visdrone_sliced
|
||||
|
||||
EvalDataset:
|
||||
!COCODataSet
|
||||
image_dir: val_images_640_025
|
||||
anno_path: val_640_025.json
|
||||
dataset_dir: dataset/visdrone_sliced
|
||||
worker_num: 0
|
||||
|
||||
# preprocess reader in test
|
||||
EvalReader:
|
||||
sample_transforms:
|
||||
- Decode: {}
|
||||
- Resize: {target_size: [640, 640], keep_ratio: False, interp: 2}
|
||||
#- NormalizeImage: {mean: [0., 0., 0.], std: [1., 1., 1.], norm_type: none}
|
||||
- Permute: {}
|
||||
batch_size: 16
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/ppyoloe_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./ppyoloe_crn_l_300e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
use_pact: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 5000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 6000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/ppyoloe_plus_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ../../output_inference/ppyoloe_plus_crn_t_auxhead_300e_coco/
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
onnx_format: True
|
||||
use_pact: False
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 8000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 6000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/ppyoloe_plus_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./ppyoloe_plus_crn_l_80e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
use_pact: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 5000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 6000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/ppyoloe_plus_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./ppyoloe_plus_crn_m_80e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
use_pact: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 5000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 6000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
|
||||
@@ -0,0 +1,26 @@
|
||||
metric: COCO
|
||||
num_classes: 80
|
||||
|
||||
# Datset configuration
|
||||
TrainDataset:
|
||||
!COCODataSet
|
||||
image_dir: train2017
|
||||
anno_path: annotations/instances_train2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
EvalDataset:
|
||||
!COCODataSet
|
||||
image_dir: val2017
|
||||
anno_path: annotations/instances_val2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
worker_num: 0
|
||||
|
||||
# preprocess reader in test
|
||||
EvalReader:
|
||||
sample_transforms:
|
||||
- Decode: {}
|
||||
- Resize: {target_size: [640, 640], keep_ratio: False, interp: 2}
|
||||
- NormalizeImage: {mean: [0., 0., 0.], std: [1., 1., 1.], norm_type: none}
|
||||
- Permute: {}
|
||||
batch_size: 4
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/ppyoloe_plus_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./ppyoloe_plus_crn_s_80e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
use_pact: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 5000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 6000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
|
||||
@@ -0,0 +1,33 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/ppyoloe_plus_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ../../output_inference/ppyoloe_plus_sod_crn_l_80e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
onnx_format: True
|
||||
use_pact: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 1
|
||||
eval_iter: 1
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 6000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/ppyoloe_plus_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./ppyoloe_plus_crn_x_80e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
use_pact: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 5000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 6000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
|
||||
@@ -0,0 +1,26 @@
|
||||
metric: COCO
|
||||
num_classes: 80
|
||||
|
||||
# Datset configuration
|
||||
TrainDataset:
|
||||
!COCODataSet
|
||||
image_dir: train2017
|
||||
anno_path: annotations/instances_train2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
EvalDataset:
|
||||
!COCODataSet
|
||||
image_dir: val2017
|
||||
anno_path: annotations/instances_val2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
worker_num: 0
|
||||
|
||||
# preprocess reader in test
|
||||
EvalReader:
|
||||
sample_transforms:
|
||||
- Decode: {}
|
||||
- Resize: {target_size: [640, 640], keep_ratio: False, interp: 2}
|
||||
- NormalizeImage: {mean: [0.485, 0.456, 0.406], std: [0.229, 0.224, 0.225], is_scale: True}
|
||||
- Permute: {}
|
||||
batch_size: 4
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/rtdetr_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./rtdetr_hgnetv2_l_6x_coco/
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
onnx_format: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
- matmul_v2
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 200
|
||||
eval_iter: 50
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 10000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/rtdetr_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./rtdetr_hgnetv2_x_6x_coco/
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
onnx_format: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
- matmul_v2
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 500
|
||||
eval_iter: 100
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 10000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/rtdetr_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./rtdetr_r101vd_6x_coco/
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
onnx_format: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
- matmul_v2
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 200
|
||||
eval_iter: 50
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 10000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/rtdetr_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./rtdetr_r50vd_6x_coco/
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
onnx_format: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
- matmul_v2
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 500
|
||||
eval_iter: 100
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 10000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
@@ -0,0 +1,38 @@
|
||||
metric: COCO
|
||||
num_classes: 80
|
||||
|
||||
# Datset configuration
|
||||
TrainDataset:
|
||||
!COCODataSet
|
||||
image_dir: train2017
|
||||
anno_path: annotations/instances_train2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
!COCODataSet
|
||||
image_dir: val2017
|
||||
anno_path: annotations/instances_val2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
EvalDataset:
|
||||
!COCODataSet
|
||||
image_dir: val2017
|
||||
anno_path: annotations/instances_val2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
TestDataset:
|
||||
!COCODataSet
|
||||
image_dir: val2017
|
||||
anno_path: annotations/instances_val2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
worker_num: 0
|
||||
|
||||
# preprocess reader in test
|
||||
EvalReader:
|
||||
sample_transforms:
|
||||
- Decode: {}
|
||||
- Resize: {target_size: [640, 640], keep_ratio: False, interp: 2}
|
||||
- NormalizeImage: {mean: [0., 0., 0.], std: [1., 1., 1.], norm_type: none}
|
||||
- Permute: {}
|
||||
batch_size: 1
|
||||
shuffle: false
|
||||
drop_last: false
|
||||
@@ -0,0 +1,26 @@
|
||||
metric: COCO
|
||||
num_classes: 80
|
||||
|
||||
# Datset configuration
|
||||
TrainDataset:
|
||||
!COCODataSet
|
||||
image_dir: train2017
|
||||
anno_path: annotations/instances_train2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
EvalDataset:
|
||||
!COCODataSet
|
||||
image_dir: val2017
|
||||
anno_path: annotations/instances_val2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
worker_num: 0
|
||||
|
||||
# preprocess reader in test
|
||||
TestReader:
|
||||
sample_transforms:
|
||||
- Decode: {}
|
||||
- Resize: {target_size: [640, 640], keep_ratio: True, interp: 1}
|
||||
- Pad: {size: [640, 640], fill_value: [114., 114., 114.]}
|
||||
- Permute: {}
|
||||
batch_size: 1
|
||||
@@ -0,0 +1,29 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/yolov5_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./yolov5_s_300e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
use_pact: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 3000
|
||||
eval_iter: 1000
|
||||
learning_rate: 0.00001
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
target_metric: 0.365
|
||||
@@ -0,0 +1,30 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/yolov5_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./yolov6mt_s_400e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 8000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 8000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 0.00004
|
||||
@@ -0,0 +1,30 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/yolov5_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./yolov7_l_300e_coco
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 8000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 8000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 0.00004
|
||||
@@ -0,0 +1,27 @@
|
||||
metric: COCO
|
||||
num_classes: 80
|
||||
|
||||
# Dataset configuration
|
||||
TrainDataset:
|
||||
!COCODataSet
|
||||
image_dir: train2017
|
||||
anno_path: annotations/instances_train2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
EvalDataset:
|
||||
!COCODataSet
|
||||
image_dir: val2017
|
||||
anno_path: annotations/instances_val2017.json
|
||||
dataset_dir: dataset/coco/
|
||||
|
||||
worker_num: 0
|
||||
|
||||
# preprocess reader in test
|
||||
EvalReader:
|
||||
sample_transforms:
|
||||
- Decode: {}
|
||||
- Resize: {target_size: [640, 640], keep_ratio: True, interp: 1}
|
||||
- Pad: {size: [640, 640], fill_value: [114., 114., 114.]}
|
||||
- NormalizeImage: {mean: [0., 0., 0.], std: [1., 1., 1.], norm_type: none}
|
||||
- Permute: {}
|
||||
batch_size: 4
|
||||
@@ -0,0 +1,32 @@
|
||||
|
||||
Global:
|
||||
reader_config: configs/yolov8_reader.yml
|
||||
include_nms: True
|
||||
Evaluation: True
|
||||
model_dir: ./yolov8_s_500e_coco_trt_nms/
|
||||
model_filename: model.pdmodel
|
||||
params_filename: model.pdiparams
|
||||
|
||||
Distillation:
|
||||
alpha: 1.0
|
||||
loss: soft_label
|
||||
|
||||
QuantAware:
|
||||
onnx_format: true
|
||||
activation_quantize_type: 'moving_average_abs_max'
|
||||
quantize_op_types:
|
||||
- conv2d
|
||||
- depthwise_conv2d
|
||||
|
||||
TrainConfig:
|
||||
train_iter: 8000
|
||||
eval_iter: 1000
|
||||
learning_rate:
|
||||
type: CosineAnnealingDecay
|
||||
learning_rate: 0.00003
|
||||
T_max: 10000
|
||||
optimizer_builder:
|
||||
optimizer:
|
||||
type: SGD
|
||||
weight_decay: 4.0e-05
|
||||
|
||||
163
paddle_detection/deploy/auto_compression/eval.py
Normal file
163
paddle_detection/deploy/auto_compression/eval.py
Normal file
@@ -0,0 +1,163 @@
|
||||
# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
import os
|
||||
import sys
|
||||
import numpy as np
|
||||
import argparse
|
||||
import paddle
|
||||
from ppdet.core.workspace import load_config, merge_config
|
||||
from ppdet.core.workspace import create
|
||||
from ppdet.metrics import COCOMetric, VOCMetric, KeyPointTopDownCOCOEval
|
||||
from paddleslim.auto_compression.config_helpers import load_config as load_slim_config
|
||||
from post_process import PPYOLOEPostProcess
|
||||
|
||||
|
||||
def argsparser():
|
||||
parser = argparse.ArgumentParser(description=__doc__)
|
||||
parser.add_argument(
|
||||
'--config_path',
|
||||
type=str,
|
||||
default=None,
|
||||
help="path of compression strategy config.",
|
||||
required=True)
|
||||
parser.add_argument(
|
||||
'--devices',
|
||||
type=str,
|
||||
default='gpu',
|
||||
help="which device used to compress.")
|
||||
|
||||
return parser
|
||||
|
||||
|
||||
def reader_wrapper(reader, input_list):
|
||||
def gen():
|
||||
for data in reader:
|
||||
in_dict = {}
|
||||
if isinstance(input_list, list):
|
||||
for input_name in input_list:
|
||||
in_dict[input_name] = data[input_name]
|
||||
elif isinstance(input_list, dict):
|
||||
for input_name in input_list.keys():
|
||||
in_dict[input_list[input_name]] = data[input_name]
|
||||
yield in_dict
|
||||
|
||||
return gen
|
||||
|
||||
|
||||
def convert_numpy_data(data, metric):
|
||||
data_all = {}
|
||||
data_all = {k: np.array(v) for k, v in data.items()}
|
||||
if isinstance(metric, VOCMetric):
|
||||
for k, v in data_all.items():
|
||||
if not isinstance(v[0], np.ndarray):
|
||||
tmp_list = []
|
||||
for t in v:
|
||||
tmp_list.append(np.array(t))
|
||||
data_all[k] = np.array(tmp_list)
|
||||
else:
|
||||
data_all = {k: np.array(v) for k, v in data.items()}
|
||||
return data_all
|
||||
|
||||
|
||||
def eval():
|
||||
|
||||
place = paddle.CUDAPlace(0) if FLAGS.devices == 'gpu' else paddle.CPUPlace()
|
||||
exe = paddle.static.Executor(place)
|
||||
|
||||
val_program, feed_target_names, fetch_targets = paddle.static.load_inference_model(
|
||||
global_config["model_dir"].rstrip('/'),
|
||||
exe,
|
||||
model_filename=global_config["model_filename"],
|
||||
params_filename=global_config["params_filename"])
|
||||
print('Loaded model from: {}'.format(global_config["model_dir"]))
|
||||
|
||||
metric = global_config['metric']
|
||||
for batch_id, data in enumerate(val_loader):
|
||||
data_all = convert_numpy_data(data, metric)
|
||||
data_input = {}
|
||||
for k, v in data.items():
|
||||
if isinstance(global_config['input_list'], list):
|
||||
if k in global_config['input_list']:
|
||||
data_input[k] = np.array(v)
|
||||
elif isinstance(global_config['input_list'], dict):
|
||||
if k in global_config['input_list'].keys():
|
||||
data_input[global_config['input_list'][k]] = np.array(v)
|
||||
|
||||
outs = exe.run(val_program,
|
||||
feed=data_input,
|
||||
fetch_list=fetch_targets,
|
||||
return_numpy=False)
|
||||
res = {}
|
||||
if 'arch' in global_config and global_config['arch'] == 'PPYOLOE':
|
||||
postprocess = PPYOLOEPostProcess(
|
||||
score_threshold=0.01, nms_threshold=0.6)
|
||||
res = postprocess(np.array(outs[0]), data_all['scale_factor'])
|
||||
else:
|
||||
for out in outs:
|
||||
v = np.array(out)
|
||||
if len(v.shape) > 1:
|
||||
res['bbox'] = v
|
||||
else:
|
||||
res['bbox_num'] = v
|
||||
metric.update(data_all, res)
|
||||
if batch_id % 100 == 0:
|
||||
print('Eval iter:', batch_id)
|
||||
metric.accumulate()
|
||||
metric.log()
|
||||
metric.reset()
|
||||
|
||||
|
||||
def main():
|
||||
global global_config
|
||||
all_config = load_slim_config(FLAGS.config_path)
|
||||
assert "Global" in all_config, "Key 'Global' not found in config file."
|
||||
global_config = all_config["Global"]
|
||||
reader_cfg = load_config(global_config['reader_config'])
|
||||
|
||||
dataset = reader_cfg['EvalDataset']
|
||||
global val_loader
|
||||
val_loader = create('EvalReader')(reader_cfg['EvalDataset'],
|
||||
reader_cfg['worker_num'],
|
||||
return_list=True)
|
||||
metric = None
|
||||
if reader_cfg['metric'] == 'COCO':
|
||||
clsid2catid = {v: k for k, v in dataset.catid2clsid.items()}
|
||||
anno_file = dataset.get_anno()
|
||||
metric = COCOMetric(
|
||||
anno_file=anno_file, clsid2catid=clsid2catid, IouType='bbox')
|
||||
elif reader_cfg['metric'] == 'VOC':
|
||||
metric = VOCMetric(
|
||||
label_list=dataset.get_label_list(),
|
||||
class_num=reader_cfg['num_classes'],
|
||||
map_type=reader_cfg['map_type'])
|
||||
elif reader_cfg['metric'] == 'KeyPointTopDownCOCOEval':
|
||||
anno_file = dataset.get_anno()
|
||||
metric = KeyPointTopDownCOCOEval(anno_file,
|
||||
len(dataset), 17, 'output_eval')
|
||||
else:
|
||||
raise ValueError("metric currently only supports COCO and VOC.")
|
||||
global_config['metric'] = metric
|
||||
|
||||
eval()
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
paddle.enable_static()
|
||||
parser = argsparser()
|
||||
FLAGS = parser.parse_args()
|
||||
assert FLAGS.devices in ['cpu', 'gpu', 'xpu', 'npu']
|
||||
paddle.set_device(FLAGS.devices)
|
||||
|
||||
main()
|
||||
@@ -0,0 +1,446 @@
|
||||
#opyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
import os
|
||||
import argparse
|
||||
import time
|
||||
import sys
|
||||
import cv2
|
||||
import numpy as np
|
||||
|
||||
import paddle
|
||||
from paddle.inference import Config
|
||||
from paddle.inference import create_predictor
|
||||
from ppdet.core.workspace import load_config, create
|
||||
from ppdet.metrics import COCOMetric
|
||||
|
||||
from post_process import PPYOLOEPostProcess
|
||||
|
||||
|
||||
def argsparser():
|
||||
"""
|
||||
argsparser func
|
||||
"""
|
||||
parser = argparse.ArgumentParser()
|
||||
parser.add_argument(
|
||||
"--model_path", type=str, help="inference model filepath")
|
||||
parser.add_argument(
|
||||
"--image_file",
|
||||
type=str,
|
||||
default=None,
|
||||
help="image path, if set image_file, it will not eval coco.")
|
||||
parser.add_argument(
|
||||
"--reader_config",
|
||||
type=str,
|
||||
default=None,
|
||||
help="path of datset and reader config.")
|
||||
parser.add_argument(
|
||||
"--benchmark",
|
||||
type=bool,
|
||||
default=False,
|
||||
help="Whether run benchmark or not.")
|
||||
parser.add_argument(
|
||||
"--use_trt",
|
||||
type=bool,
|
||||
default=False,
|
||||
help="Whether use TensorRT or not.")
|
||||
parser.add_argument(
|
||||
"--precision",
|
||||
type=str,
|
||||
default="paddle",
|
||||
help="mode of running(fp32/fp16/int8)")
|
||||
parser.add_argument(
|
||||
"--device",
|
||||
type=str,
|
||||
default="GPU",
|
||||
help="Choose the device you want to run, it can be: CPU/GPU/XPU, default is GPU",
|
||||
)
|
||||
parser.add_argument(
|
||||
"--use_dynamic_shape",
|
||||
type=bool,
|
||||
default=True,
|
||||
help="Whether use dynamic shape or not.")
|
||||
parser.add_argument(
|
||||
"--use_mkldnn",
|
||||
type=bool,
|
||||
default=False,
|
||||
help="Whether use mkldnn or not.")
|
||||
parser.add_argument(
|
||||
"--cpu_threads", type=int, default=10, help="Num of cpu threads.")
|
||||
parser.add_argument("--img_shape", type=int, default=640, help="input_size")
|
||||
parser.add_argument(
|
||||
'--include_nms',
|
||||
type=bool,
|
||||
default=True,
|
||||
help="Whether include nms or not.")
|
||||
|
||||
return parser
|
||||
|
||||
|
||||
CLASS_LABEL = [
|
||||
'person', 'bicycle', 'car', 'motorcycle', 'airplane', 'bus', 'train',
|
||||
'truck', 'boat', 'traffic light', 'fire hydrant', 'stop sign',
|
||||
'parking meter', 'bench', 'bird', 'cat', 'dog', 'horse', 'sheep', 'cow',
|
||||
'elephant', 'bear', 'zebra', 'giraffe', 'backpack', 'umbrella', 'handbag',
|
||||
'tie', 'suitcase', 'frisbee', 'skis', 'snowboard', 'sports ball', 'kite',
|
||||
'baseball bat', 'baseball glove', 'skateboard', 'surfboard',
|
||||
'tennis racket', 'bottle', 'wine glass', 'cup', 'fork', 'knife', 'spoon',
|
||||
'bowl', 'banana', 'apple', 'sandwich', 'orange', 'broccoli', 'carrot',
|
||||
'hot dog', 'pizza', 'donut', 'cake', 'chair', 'couch', 'potted plant',
|
||||
'bed', 'dining table', 'toilet', 'tv', 'laptop', 'mouse', 'remote',
|
||||
'keyboard', 'cell phone', 'microwave', 'oven', 'toaster', 'sink',
|
||||
'refrigerator', 'book', 'clock', 'vase', 'scissors', 'teddy bear',
|
||||
'hair drier', 'toothbrush'
|
||||
]
|
||||
|
||||
|
||||
def generate_scale(im, target_shape, keep_ratio=True):
|
||||
"""
|
||||
Args:
|
||||
im (np.ndarray): image (np.ndarray)
|
||||
Returns:
|
||||
im_scale_x: the resize ratio of X
|
||||
im_scale_y: the resize ratio of Y
|
||||
"""
|
||||
origin_shape = im.shape[:2]
|
||||
if keep_ratio:
|
||||
im_size_min = np.min(origin_shape)
|
||||
im_size_max = np.max(origin_shape)
|
||||
target_size_min = np.min(target_shape)
|
||||
target_size_max = np.max(target_shape)
|
||||
im_scale = float(target_size_min) / float(im_size_min)
|
||||
if np.round(im_scale * im_size_max) > target_size_max:
|
||||
im_scale = float(target_size_max) / float(im_size_max)
|
||||
im_scale_x = im_scale
|
||||
im_scale_y = im_scale
|
||||
else:
|
||||
resize_h, resize_w = target_shape
|
||||
im_scale_y = resize_h / float(origin_shape[0])
|
||||
im_scale_x = resize_w / float(origin_shape[1])
|
||||
return im_scale_y, im_scale_x
|
||||
|
||||
|
||||
def image_preprocess(img_path, target_shape):
|
||||
"""
|
||||
image_preprocess func
|
||||
"""
|
||||
img = cv2.imread(img_path)
|
||||
im_scale_y, im_scale_x = generate_scale(img, target_shape, keep_ratio=False)
|
||||
img = cv2.resize(
|
||||
img, (target_shape[0], target_shape[0]),
|
||||
interpolation=cv2.INTER_LANCZOS4)
|
||||
img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
|
||||
img = np.transpose(img, [2, 0, 1]) / 255
|
||||
img = np.expand_dims(img, 0)
|
||||
img_mean = np.array([0.485, 0.456, 0.406]).reshape((3, 1, 1))
|
||||
img_std = np.array([0.229, 0.224, 0.225]).reshape((3, 1, 1))
|
||||
img -= img_mean
|
||||
img /= img_std
|
||||
scale_factor = np.array([[im_scale_y, im_scale_x]])
|
||||
return img.astype(np.float32), scale_factor.astype(np.float32)
|
||||
|
||||
|
||||
def get_color_map_list(num_classes):
|
||||
"""
|
||||
get_color_map_list func
|
||||
"""
|
||||
color_map = num_classes * [0, 0, 0]
|
||||
for i in range(0, num_classes):
|
||||
j = 0
|
||||
lab = i
|
||||
while lab:
|
||||
color_map[i * 3] |= ((lab >> 0) & 1) << (7 - j)
|
||||
color_map[i * 3 + 1] |= ((lab >> 1) & 1) << (7 - j)
|
||||
color_map[i * 3 + 2] |= ((lab >> 2) & 1) << (7 - j)
|
||||
j += 1
|
||||
lab >>= 3
|
||||
color_map = [color_map[i:i + 3] for i in range(0, len(color_map), 3)]
|
||||
return color_map
|
||||
|
||||
|
||||
def draw_box(image_file, results, class_label, threshold=0.5):
|
||||
"""
|
||||
draw_box func
|
||||
"""
|
||||
srcimg = cv2.imread(image_file, 1)
|
||||
for i in range(len(results)):
|
||||
color_list = get_color_map_list(len(class_label))
|
||||
clsid2color = {}
|
||||
classid, conf = int(results[i, 0]), results[i, 1]
|
||||
if conf < threshold:
|
||||
continue
|
||||
xmin, ymin, xmax, ymax = int(results[i, 2]), int(results[i, 3]), int(
|
||||
results[i, 4]), int(results[i, 5])
|
||||
|
||||
if classid not in clsid2color:
|
||||
clsid2color[classid] = color_list[classid]
|
||||
color = tuple(clsid2color[classid])
|
||||
|
||||
cv2.rectangle(srcimg, (xmin, ymin), (xmax, ymax), color, thickness=2)
|
||||
print(class_label[classid] + ": " + str(round(conf, 3)))
|
||||
cv2.putText(
|
||||
srcimg,
|
||||
class_label[classid] + ":" + str(round(conf, 3)),
|
||||
(xmin, ymin - 10),
|
||||
cv2.FONT_HERSHEY_SIMPLEX,
|
||||
0.8,
|
||||
(0, 255, 0),
|
||||
thickness=2, )
|
||||
return srcimg
|
||||
|
||||
|
||||
def load_predictor(
|
||||
model_dir,
|
||||
precision="fp32",
|
||||
use_trt=False,
|
||||
use_mkldnn=False,
|
||||
batch_size=1,
|
||||
device="CPU",
|
||||
min_subgraph_size=3,
|
||||
use_dynamic_shape=False,
|
||||
trt_min_shape=1,
|
||||
trt_max_shape=1280,
|
||||
trt_opt_shape=640,
|
||||
cpu_threads=1, ):
|
||||
"""set AnalysisConfig, generate AnalysisPredictor
|
||||
Args:
|
||||
model_dir (str): root path of __model__ and __params__
|
||||
precision (str): mode of running(fp32/fp16/int8)
|
||||
use_trt (bool): whether use TensorRT or not.
|
||||
use_mkldnn (bool): whether use MKLDNN or not in CPU.
|
||||
device (str): Choose the device you want to run, it can be: CPU/GPU, default is CPU
|
||||
use_dynamic_shape (bool): use dynamic shape or not
|
||||
trt_min_shape (int): min shape for dynamic shape in trt
|
||||
trt_max_shape (int): max shape for dynamic shape in trt
|
||||
trt_opt_shape (int): opt shape for dynamic shape in trt
|
||||
Returns:
|
||||
predictor (PaddlePredictor): AnalysisPredictor
|
||||
Raises:
|
||||
ValueError: predict by TensorRT need device == 'GPU'.
|
||||
"""
|
||||
rerun_flag = False
|
||||
if device != "GPU" and use_trt:
|
||||
raise ValueError(
|
||||
"Predict by TensorRT mode: {}, expect device=='GPU', but device == {}".
|
||||
format(precision, device))
|
||||
config = Config(
|
||||
os.path.join(model_dir, "model.pdmodel"),
|
||||
os.path.join(model_dir, "model.pdiparams"))
|
||||
if device == "GPU":
|
||||
# initial GPU memory(M), device ID
|
||||
config.enable_use_gpu(200, 0)
|
||||
# optimize graph and fuse op
|
||||
config.switch_ir_optim(True)
|
||||
else:
|
||||
config.disable_gpu()
|
||||
config.set_cpu_math_library_num_threads(cpu_threads)
|
||||
config.switch_ir_optim()
|
||||
if use_mkldnn:
|
||||
config.enable_mkldnn()
|
||||
if precision == "int8":
|
||||
config.enable_mkldnn_int8(
|
||||
{"conv2d", "depthwise_conv2d", "transpose2", "pool2d"})
|
||||
|
||||
precision_map = {
|
||||
"int8": Config.Precision.Int8,
|
||||
"fp32": Config.Precision.Float32,
|
||||
"fp16": Config.Precision.Half,
|
||||
}
|
||||
if precision in precision_map.keys() and use_trt:
|
||||
config.enable_tensorrt_engine(
|
||||
workspace_size=(1 << 25) * batch_size,
|
||||
max_batch_size=batch_size,
|
||||
min_subgraph_size=min_subgraph_size,
|
||||
precision_mode=precision_map[precision],
|
||||
use_static=True,
|
||||
use_calib_mode=False, )
|
||||
|
||||
if use_dynamic_shape:
|
||||
dynamic_shape_file = os.path.join(FLAGS.model_path,
|
||||
"dynamic_shape.txt")
|
||||
if os.path.exists(dynamic_shape_file):
|
||||
config.enable_tuned_tensorrt_dynamic_shape(dynamic_shape_file,
|
||||
True)
|
||||
print("trt set dynamic shape done!")
|
||||
else:
|
||||
config.collect_shape_range_info(dynamic_shape_file)
|
||||
print("Start collect dynamic shape...")
|
||||
rerun_flag = True
|
||||
|
||||
# enable shared memory
|
||||
config.enable_memory_optim()
|
||||
predictor = create_predictor(config)
|
||||
return predictor, rerun_flag
|
||||
|
||||
|
||||
def predict_image(predictor,
|
||||
image_file,
|
||||
image_shape=[640, 640],
|
||||
warmup=1,
|
||||
repeats=1,
|
||||
threshold=0.5):
|
||||
"""
|
||||
predict image main func
|
||||
"""
|
||||
img, scale_factor = image_preprocess(image_file, image_shape)
|
||||
inputs = {}
|
||||
inputs["image"] = img
|
||||
if FLAGS.include_nms:
|
||||
inputs['scale_factor'] = scale_factor
|
||||
input_names = predictor.get_input_names()
|
||||
for i, _ in enumerate(input_names):
|
||||
input_tensor = predictor.get_input_handle(input_names[i])
|
||||
input_tensor.copy_from_cpu(inputs[input_names[i]])
|
||||
|
||||
for i in range(warmup):
|
||||
predictor.run()
|
||||
|
||||
np_boxes, np_boxes_num = None, None
|
||||
cpu_mems, gpu_mems = 0, 0
|
||||
predict_time = 0.0
|
||||
time_min = float("inf")
|
||||
time_max = float("-inf")
|
||||
for i in range(repeats):
|
||||
start_time = time.time()
|
||||
predictor.run()
|
||||
output_names = predictor.get_output_names()
|
||||
boxes_tensor = predictor.get_output_handle(output_names[0])
|
||||
np_boxes = boxes_tensor.copy_to_cpu()
|
||||
if FLAGS.include_nms:
|
||||
boxes_num = predictor.get_output_handle(output_names[1])
|
||||
np_boxes_num = boxes_num.copy_to_cpu()
|
||||
end_time = time.time()
|
||||
timed = end_time - start_time
|
||||
time_min = min(time_min, timed)
|
||||
time_max = max(time_max, timed)
|
||||
predict_time += timed
|
||||
time_avg = predict_time / repeats
|
||||
print("[Benchmark]Inference time(ms): min={}, max={}, avg={}".format(
|
||||
round(time_min * 1000, 2),
|
||||
round(time_max * 1000, 1), round(time_avg * 1000, 1)))
|
||||
if not FLAGS.include_nms:
|
||||
postprocess = PPYOLOEPostProcess(score_threshold=0.3, nms_threshold=0.6)
|
||||
res = postprocess(np_boxes, scale_factor)
|
||||
else:
|
||||
res = {'bbox': np_boxes, 'bbox_num': np_boxes_num}
|
||||
res_img = draw_box(
|
||||
image_file, res["bbox"], CLASS_LABEL, threshold=threshold)
|
||||
cv2.imwrite("result.jpg", res_img)
|
||||
|
||||
|
||||
def eval(predictor, val_loader, metric, rerun_flag=False):
|
||||
"""
|
||||
eval main func
|
||||
"""
|
||||
cpu_mems, gpu_mems = 0, 0
|
||||
predict_time = 0.0
|
||||
time_min = float("inf")
|
||||
time_max = float("-inf")
|
||||
sample_nums = len(val_loader)
|
||||
input_names = predictor.get_input_names()
|
||||
output_names = predictor.get_output_names()
|
||||
boxes_tensor = predictor.get_output_handle(output_names[0])
|
||||
if FLAGS.include_nms:
|
||||
boxes_num = predictor.get_output_handle(output_names[1])
|
||||
for batch_id, data in enumerate(val_loader):
|
||||
data_all = {k: np.array(v) for k, v in data.items()}
|
||||
for i, _ in enumerate(input_names):
|
||||
input_tensor = predictor.get_input_handle(input_names[i])
|
||||
input_tensor.copy_from_cpu(data_all[input_names[i]])
|
||||
start_time = time.time()
|
||||
predictor.run()
|
||||
np_boxes = boxes_tensor.copy_to_cpu()
|
||||
if FLAGS.include_nms:
|
||||
np_boxes_num = boxes_num.copy_to_cpu()
|
||||
if rerun_flag:
|
||||
return
|
||||
end_time = time.time()
|
||||
timed = end_time - start_time
|
||||
time_min = min(time_min, timed)
|
||||
time_max = max(time_max, timed)
|
||||
predict_time += timed
|
||||
if not FLAGS.include_nms:
|
||||
postprocess = PPYOLOEPostProcess(
|
||||
score_threshold=0.3, nms_threshold=0.6)
|
||||
res = postprocess(np_boxes, data_all['scale_factor'])
|
||||
else:
|
||||
res = {'bbox': np_boxes, 'bbox_num': np_boxes_num}
|
||||
metric.update(data_all, res)
|
||||
if batch_id % 100 == 0:
|
||||
print("Eval iter:", batch_id)
|
||||
sys.stdout.flush()
|
||||
metric.accumulate()
|
||||
metric.log()
|
||||
map_res = metric.get_results()
|
||||
metric.reset()
|
||||
time_avg = predict_time / sample_nums
|
||||
print("[Benchmark]Inference time(ms): min={}, max={}, avg={}".format(
|
||||
round(time_min * 1000, 2),
|
||||
round(time_max * 1000, 1), round(time_avg * 1000, 1)))
|
||||
print("[Benchmark] COCO mAP: {}".format(map_res["bbox"][0]))
|
||||
sys.stdout.flush()
|
||||
|
||||
|
||||
def main():
|
||||
"""
|
||||
main func
|
||||
"""
|
||||
predictor, rerun_flag = load_predictor(
|
||||
FLAGS.model_path,
|
||||
device=FLAGS.device,
|
||||
use_trt=FLAGS.use_trt,
|
||||
use_mkldnn=FLAGS.use_mkldnn,
|
||||
precision=FLAGS.precision,
|
||||
use_dynamic_shape=FLAGS.use_dynamic_shape,
|
||||
cpu_threads=FLAGS.cpu_threads)
|
||||
|
||||
if FLAGS.image_file:
|
||||
warmup, repeats = 1, 1
|
||||
if FLAGS.benchmark:
|
||||
warmup, repeats = 50, 100
|
||||
predict_image(
|
||||
predictor,
|
||||
FLAGS.image_file,
|
||||
image_shape=[FLAGS.img_shape, FLAGS.img_shape],
|
||||
warmup=warmup,
|
||||
repeats=repeats)
|
||||
else:
|
||||
reader_cfg = load_config(FLAGS.reader_config)
|
||||
|
||||
dataset = reader_cfg["EvalDataset"]
|
||||
global val_loader
|
||||
val_loader = create("EvalReader")(reader_cfg["EvalDataset"],
|
||||
reader_cfg["worker_num"],
|
||||
return_list=True)
|
||||
clsid2catid = {v: k for k, v in dataset.catid2clsid.items()}
|
||||
anno_file = dataset.get_anno()
|
||||
metric = COCOMetric(
|
||||
anno_file=anno_file, clsid2catid=clsid2catid, IouType="bbox")
|
||||
eval(predictor, val_loader, metric, rerun_flag=rerun_flag)
|
||||
|
||||
if rerun_flag:
|
||||
print(
|
||||
"***** Collect dynamic shape done, Please rerun the program to get correct results. *****"
|
||||
)
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
paddle.enable_static()
|
||||
parser = argsparser()
|
||||
FLAGS = parser.parse_args()
|
||||
|
||||
# DataLoader need run on cpu
|
||||
paddle.set_device("cpu")
|
||||
|
||||
main()
|
||||
157
paddle_detection/deploy/auto_compression/post_process.py
Normal file
157
paddle_detection/deploy/auto_compression/post_process.py
Normal file
@@ -0,0 +1,157 @@
|
||||
# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
import numpy as np
|
||||
import cv2
|
||||
|
||||
|
||||
def hard_nms(box_scores, iou_threshold, top_k=-1, candidate_size=200):
|
||||
"""
|
||||
Args:
|
||||
box_scores (N, 5): boxes in corner-form and probabilities.
|
||||
iou_threshold: intersection over union threshold.
|
||||
top_k: keep top_k results. If k <= 0, keep all the results.
|
||||
candidate_size: only consider the candidates with the highest scores.
|
||||
Returns:
|
||||
picked: a list of indexes of the kept boxes
|
||||
"""
|
||||
scores = box_scores[:, -1]
|
||||
boxes = box_scores[:, :-1]
|
||||
picked = []
|
||||
indexes = np.argsort(scores)
|
||||
indexes = indexes[-candidate_size:]
|
||||
while len(indexes) > 0:
|
||||
current = indexes[-1]
|
||||
picked.append(current)
|
||||
if 0 < top_k == len(picked) or len(indexes) == 1:
|
||||
break
|
||||
current_box = boxes[current, :]
|
||||
indexes = indexes[:-1]
|
||||
rest_boxes = boxes[indexes, :]
|
||||
iou = iou_of(
|
||||
rest_boxes,
|
||||
np.expand_dims(
|
||||
current_box, axis=0), )
|
||||
indexes = indexes[iou <= iou_threshold]
|
||||
|
||||
return box_scores[picked, :]
|
||||
|
||||
|
||||
def iou_of(boxes0, boxes1, eps=1e-5):
|
||||
"""Return intersection-over-union (Jaccard index) of boxes.
|
||||
Args:
|
||||
boxes0 (N, 4): ground truth boxes.
|
||||
boxes1 (N or 1, 4): predicted boxes.
|
||||
eps: a small number to avoid 0 as denominator.
|
||||
Returns:
|
||||
iou (N): IoU values.
|
||||
"""
|
||||
overlap_left_top = np.maximum(boxes0[..., :2], boxes1[..., :2])
|
||||
overlap_right_bottom = np.minimum(boxes0[..., 2:], boxes1[..., 2:])
|
||||
|
||||
overlap_area = area_of(overlap_left_top, overlap_right_bottom)
|
||||
area0 = area_of(boxes0[..., :2], boxes0[..., 2:])
|
||||
area1 = area_of(boxes1[..., :2], boxes1[..., 2:])
|
||||
return overlap_area / (area0 + area1 - overlap_area + eps)
|
||||
|
||||
|
||||
def area_of(left_top, right_bottom):
|
||||
"""Compute the areas of rectangles given two corners.
|
||||
Args:
|
||||
left_top (N, 2): left top corner.
|
||||
right_bottom (N, 2): right bottom corner.
|
||||
Returns:
|
||||
area (N): return the area.
|
||||
"""
|
||||
hw = np.clip(right_bottom - left_top, 0.0, None)
|
||||
return hw[..., 0] * hw[..., 1]
|
||||
|
||||
|
||||
class PPYOLOEPostProcess(object):
|
||||
"""
|
||||
Args:
|
||||
input_shape (int): network input image size
|
||||
scale_factor (float): scale factor of ori image
|
||||
"""
|
||||
|
||||
def __init__(self,
|
||||
score_threshold=0.4,
|
||||
nms_threshold=0.5,
|
||||
nms_top_k=10000,
|
||||
keep_top_k=300):
|
||||
self.score_threshold = score_threshold
|
||||
self.nms_threshold = nms_threshold
|
||||
self.nms_top_k = nms_top_k
|
||||
self.keep_top_k = keep_top_k
|
||||
|
||||
def _non_max_suppression(self, prediction, scale_factor):
|
||||
batch_size = prediction.shape[0]
|
||||
out_boxes_list = []
|
||||
box_num_list = []
|
||||
for batch_id in range(batch_size):
|
||||
bboxes, confidences = prediction[batch_id][..., :4], prediction[
|
||||
batch_id][..., 4:]
|
||||
# nms
|
||||
picked_box_probs = []
|
||||
picked_labels = []
|
||||
for class_index in range(0, confidences.shape[1]):
|
||||
probs = confidences[:, class_index]
|
||||
mask = probs > self.score_threshold
|
||||
probs = probs[mask]
|
||||
if probs.shape[0] == 0:
|
||||
continue
|
||||
subset_boxes = bboxes[mask, :]
|
||||
box_probs = np.concatenate(
|
||||
[subset_boxes, probs.reshape(-1, 1)], axis=1)
|
||||
box_probs = hard_nms(
|
||||
box_probs,
|
||||
iou_threshold=self.nms_threshold,
|
||||
top_k=self.nms_top_k)
|
||||
picked_box_probs.append(box_probs)
|
||||
picked_labels.extend([class_index] * box_probs.shape[0])
|
||||
|
||||
if len(picked_box_probs) == 0:
|
||||
out_boxes_list.append(np.empty((0, 4)))
|
||||
|
||||
else:
|
||||
picked_box_probs = np.concatenate(picked_box_probs)
|
||||
# resize output boxes
|
||||
picked_box_probs[:, 0] /= scale_factor[batch_id][1]
|
||||
picked_box_probs[:, 2] /= scale_factor[batch_id][1]
|
||||
picked_box_probs[:, 1] /= scale_factor[batch_id][0]
|
||||
picked_box_probs[:, 3] /= scale_factor[batch_id][0]
|
||||
|
||||
# clas score box
|
||||
out_box = np.concatenate(
|
||||
[
|
||||
np.expand_dims(
|
||||
np.array(picked_labels), axis=-1), np.expand_dims(
|
||||
picked_box_probs[:, 4], axis=-1),
|
||||
picked_box_probs[:, :4]
|
||||
],
|
||||
axis=1)
|
||||
if out_box.shape[0] > self.keep_top_k:
|
||||
out_box = out_box[out_box[:, 1].argsort()[::-1]
|
||||
[:self.keep_top_k]]
|
||||
out_boxes_list.append(out_box)
|
||||
box_num_list.append(out_box.shape[0])
|
||||
|
||||
out_boxes_list = np.concatenate(out_boxes_list, axis=0)
|
||||
box_num_list = np.array(box_num_list)
|
||||
return out_boxes_list, box_num_list
|
||||
|
||||
def __call__(self, outs, scale_factor):
|
||||
out_boxes_list, box_num_list = self._non_max_suppression(outs,
|
||||
scale_factor)
|
||||
return {'bbox': out_boxes_list, 'bbox_num': box_num_list}
|
||||
191
paddle_detection/deploy/auto_compression/run.py
Normal file
191
paddle_detection/deploy/auto_compression/run.py
Normal file
@@ -0,0 +1,191 @@
|
||||
# Copyright (c) 2022 PaddlePaddle Authors. All Rights Reserved.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
import os
|
||||
import sys
|
||||
import numpy as np
|
||||
import argparse
|
||||
import paddle
|
||||
from ppdet.core.workspace import load_config, merge_config
|
||||
from ppdet.core.workspace import create
|
||||
from ppdet.metrics import COCOMetric, VOCMetric, KeyPointTopDownCOCOEval
|
||||
from paddleslim.auto_compression.config_helpers import load_config as load_slim_config
|
||||
from paddleslim.auto_compression import AutoCompression
|
||||
from post_process import PPYOLOEPostProcess
|
||||
from paddleslim.common.dataloader import get_feed_vars
|
||||
|
||||
|
||||
def argsparser():
|
||||
parser = argparse.ArgumentParser(description=__doc__)
|
||||
parser.add_argument(
|
||||
'--config_path',
|
||||
type=str,
|
||||
default=None,
|
||||
help="path of compression strategy config.",
|
||||
required=True)
|
||||
parser.add_argument(
|
||||
'--save_dir',
|
||||
type=str,
|
||||
default='output',
|
||||
help="directory to save compressed model.")
|
||||
parser.add_argument(
|
||||
'--devices',
|
||||
type=str,
|
||||
default='gpu',
|
||||
help="which device used to compress.")
|
||||
|
||||
return parser
|
||||
|
||||
|
||||
def reader_wrapper(reader, input_list):
|
||||
def gen():
|
||||
for data in reader:
|
||||
in_dict = {}
|
||||
if isinstance(input_list, list):
|
||||
for input_name in input_list:
|
||||
in_dict[input_name] = data[input_name]
|
||||
elif isinstance(input_list, dict):
|
||||
for input_name in input_list.keys():
|
||||
in_dict[input_list[input_name]] = data[input_name]
|
||||
yield in_dict
|
||||
|
||||
return gen
|
||||
|
||||
|
||||
def convert_numpy_data(data, metric):
|
||||
data_all = {}
|
||||
data_all = {k: np.array(v) for k, v in data.items()}
|
||||
if isinstance(metric, VOCMetric):
|
||||
for k, v in data_all.items():
|
||||
if not isinstance(v[0], np.ndarray):
|
||||
tmp_list = []
|
||||
for t in v:
|
||||
tmp_list.append(np.array(t))
|
||||
data_all[k] = np.array(tmp_list)
|
||||
else:
|
||||
data_all = {k: np.array(v) for k, v in data.items()}
|
||||
return data_all
|
||||
|
||||
|
||||
def eval_function(exe, compiled_test_program, test_feed_names, test_fetch_list):
|
||||
metric = global_config['metric']
|
||||
for batch_id, data in enumerate(val_loader):
|
||||
data_all = convert_numpy_data(data, metric)
|
||||
data_input = {}
|
||||
for k, v in data.items():
|
||||
if isinstance(global_config['input_list'], list):
|
||||
if k in test_feed_names:
|
||||
data_input[k] = np.array(v)
|
||||
elif isinstance(global_config['input_list'], dict):
|
||||
if k in global_config['input_list'].keys():
|
||||
data_input[global_config['input_list'][k]] = np.array(v)
|
||||
outs = exe.run(compiled_test_program,
|
||||
feed=data_input,
|
||||
fetch_list=test_fetch_list,
|
||||
return_numpy=False)
|
||||
res = {}
|
||||
if 'include_nms' in global_config and not global_config['include_nms']:
|
||||
if 'arch' in global_config and global_config['arch'] == 'PPYOLOE':
|
||||
postprocess = PPYOLOEPostProcess(
|
||||
score_threshold=0.01, nms_threshold=0.6)
|
||||
else:
|
||||
assert "Not support arch={} now.".format(global_config['arch'])
|
||||
res = postprocess(np.array(outs[0]), data_all['scale_factor'])
|
||||
else:
|
||||
for out in outs:
|
||||
v = np.array(out)
|
||||
if len(v.shape) > 1:
|
||||
res['bbox'] = v
|
||||
else:
|
||||
res['bbox_num'] = v
|
||||
|
||||
metric.update(data_all, res)
|
||||
if batch_id % 100 == 0:
|
||||
print('Eval iter:', batch_id)
|
||||
metric.accumulate()
|
||||
metric.log()
|
||||
map_res = metric.get_results()
|
||||
metric.reset()
|
||||
map_key = 'keypoint' if 'arch' in global_config and global_config[
|
||||
'arch'] == 'keypoint' else 'bbox'
|
||||
return map_res[map_key][0]
|
||||
|
||||
|
||||
def main():
|
||||
global global_config
|
||||
all_config = load_slim_config(FLAGS.config_path)
|
||||
assert "Global" in all_config, "Key 'Global' not found in config file."
|
||||
global_config = all_config["Global"]
|
||||
reader_cfg = load_config(global_config['reader_config'])
|
||||
|
||||
train_loader = create('EvalReader')(reader_cfg['TrainDataset'],
|
||||
reader_cfg['worker_num'],
|
||||
return_list=True)
|
||||
if global_config.get('input_list') is None:
|
||||
global_config['input_list'] = get_feed_vars(
|
||||
global_config['model_dir'], global_config['model_filename'],
|
||||
global_config['params_filename'])
|
||||
train_loader = reader_wrapper(train_loader, global_config['input_list'])
|
||||
|
||||
if 'Evaluation' in global_config.keys() and global_config[
|
||||
'Evaluation'] and paddle.distributed.get_rank() == 0:
|
||||
eval_func = eval_function
|
||||
dataset = reader_cfg['EvalDataset']
|
||||
global val_loader
|
||||
_eval_batch_sampler = paddle.io.BatchSampler(
|
||||
dataset, batch_size=reader_cfg['EvalReader']['batch_size'])
|
||||
val_loader = create('EvalReader')(dataset,
|
||||
reader_cfg['worker_num'],
|
||||
batch_sampler=_eval_batch_sampler,
|
||||
return_list=True)
|
||||
metric = None
|
||||
if reader_cfg['metric'] == 'COCO':
|
||||
clsid2catid = {v: k for k, v in dataset.catid2clsid.items()}
|
||||
anno_file = dataset.get_anno()
|
||||
metric = COCOMetric(
|
||||
anno_file=anno_file, clsid2catid=clsid2catid, IouType='bbox')
|
||||
elif reader_cfg['metric'] == 'VOC':
|
||||
metric = VOCMetric(
|
||||
label_list=dataset.get_label_list(),
|
||||
class_num=reader_cfg['num_classes'],
|
||||
map_type=reader_cfg['map_type'])
|
||||
elif reader_cfg['metric'] == 'KeyPointTopDownCOCOEval':
|
||||
anno_file = dataset.get_anno()
|
||||
metric = KeyPointTopDownCOCOEval(anno_file,
|
||||
len(dataset), 17, 'output_eval')
|
||||
else:
|
||||
raise ValueError("metric currently only supports COCO and VOC.")
|
||||
global_config['metric'] = metric
|
||||
else:
|
||||
eval_func = None
|
||||
|
||||
ac = AutoCompression(
|
||||
model_dir=global_config["model_dir"],
|
||||
model_filename=global_config["model_filename"],
|
||||
params_filename=global_config["params_filename"],
|
||||
save_dir=FLAGS.save_dir,
|
||||
config=all_config,
|
||||
train_dataloader=train_loader,
|
||||
eval_callback=eval_func)
|
||||
ac.compress()
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
paddle.enable_static()
|
||||
parser = argsparser()
|
||||
FLAGS = parser.parse_args()
|
||||
assert FLAGS.devices in ['cpu', 'gpu', 'xpu', 'npu']
|
||||
paddle.set_device(FLAGS.devices)
|
||||
|
||||
main()
|
||||
Reference in New Issue
Block a user