试论述外贸网站建设应注意的问题,济宁网站建设培训,电商网站费用,做企业网站用什么cms若该文为原创文章#xff0c;转载请注明原文出处。 注#xff1a;转换测试使用的是Autodl服务器#xff0c;CUDA11.1版本#xff0c;py3.8。
一、PP-YOLOE环境安装
创建环境
# 使用 conda 创建一个名为 PaddleYOLO 的环境#xff0c;并指定 python 版本conda create -n…若该文为原创文章转载请注明原文出处。 注转换测试使用的是Autodl服务器CUDA11.1版本py3.8。
一、PP-YOLOE环境安装
创建环境
# 使用 conda 创建一个名为 PaddleYOLO 的环境并指定 python 版本conda create -n PaddleYOLO python3.8
激活
conda activate PaddleYOLO
安装参考官网
# 安装 PaddlePaddleYOLO 代码库推荐使用 paddlepaddle-2.4.2 以上的版本
# 教程测试使用 conda 安装 gpu 版 paddlepaddle 2.5
python -m pip install paddlepaddle-gpu2.5.2.post112 -f https://www.paddlepaddle.org.cn/whl/linux/mkl/avx/stable.html
二、PP-YOLOE 模型简单使用
1、获取 PaddleYOLO 源码
# 拉取 PaddleYOLO
git clone https://github.com/PaddlePaddle/PaddleYOLO.git
# 切换到 PaddleYOLO 目录安装相关依赖库
cd PaddleYOLO
pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple
# install
2、模型推理
下載模型
# PP-YOLOE_s
wget https://bj.bcebos.com/v1/paddledet/models/ppyoloe_plus_crn_s_80e_coco.pdparams
# PP-YOLOE_m
wget https://bj.bcebos.com/v1/paddledet/models/ppyoloe_plus_crn_m_80e_coco.pdparams
使用 tools/infer.py 进行推理 # 可能需要安装 9.5.0 版本的 Pillow
pip install Pillow9.5.0
推理測試
python tools/infer.py -c configs/ppyoloe/ppyoloe_plus_crn_s_80e_coco.yml -o weightsppyoloe_plus_crn_s_80e_coco.pdparams --infer_imgdemo/000000014439_640x640.jpg --draw_threshold0.5
# -c 指定配置文件configs/目录下的配置文件测试使用 ppyoloe_plus_crn_s_80e_coco.yml也可以是自己添加的,
# -o 或者 --opt 设置配置选项这里设置了 weights 使用前面手动下载的权重也可以直接设置
weightshttps://bj.bcebos.com/v1/paddledet/models/ppyoloe_plus_crn_s_80e_coco.pdparams
# --infer_dir 指定推理的图片路径或者文件夹--draw_threshold 画框的阈值默认 0.5
图像尺寸是 640*640 推理正常
三、Train
1、数据集下载
# 数据集很大有18G
http://images.cocodataset.org/zips/train2017.zip
2、train
configconfigs/${model_name}/${job_name}.yml
python tools/train.py -c ${config} --eval --amp
根据readme提供的train方法
执行下面命令
python tools/train.py -c configs/ppyoloe/ppyoloe_plus_crn_s_80e_coco.yml --eval --amp
这里没有重新train,直接使用官方模型
四、板卡部署模型
1、针对 RKNN 优化
为了在 RKNPU 上获得更优的推理性能我们调整了模型的输出结构这些调整会影响到后处理
的逻辑主要包含以下内容
• DFL 结构被移至后处理• 新增额外输出该输出为所有类别分数的累加用于加速后处理的候选框过滤逻辑具体请参考 rknn_model_zoo 。
我们在前面拉取的 PaddleDetection 源码或者 PaddleYOLO 源码基础上简单修改下源码版本是release/2.6
使用的版本是
https://github.com/PaddlePaddle/PaddleYOLO/tree/release/2.6
下载后解压
cd PaddleYOLO-release-2.6
列表 1: ppdet/modeling/architectures/yolo.py
if self.training:yolo_losses self.yolo_head(neck_feats, self.inputs)return yolo_losses
else:yolo_head_outs self.yolo_head(neck_feats)return yolo_head_outs 列表 2: ppdet/modeling/heads/ppyoloe_head.py rk_out_list []
for i, feat in enumerate(feats):_, _, h, w feat.shapel h * wavg_feat F.adaptive_avg_pool2d(feat, (1, 1))cls_logit self.pred_cls[i](self.stem_cls[i](feat, avg_feat) reg_dist self.pred_reg[i](self.stem_reg[i](feat, avg_feat))rk_out_list.append(reg_dist)rk_out_list.append(F.sigmoid(cls_logit))rk_out_list.append(paddle.clip(rk_out_list[-1].sum(1, keepdimTrue), 0, 1))reg_dist reg_dist.reshape([-1, 4, self.reg_channels, l]).transpose([0, 2, 3, 1])if self.use_shared_conv:reg_dist self.proj_conv(F.softmax(reg_dist, axis1)).squeeze(1)else:reg_dist F.softmax(reg_dist, axis1)# cls and regcls_score F.sigmoid(cls_logit)cls_score_list.append(cls_score.reshape([-1, self.num_classes,1]))reg_dist_list.append(reg_dist)return rk_out_list
上面简单的修改只用于模型导出训练模型时请注释掉(上面版本是2.6不要弄错源码修改也可以直接打补丁源码版本是 release/2.5具体参考下 rknn_model_zoo/models/CV/object_detection/yolo/patch_for_model_export/ppyoloe at v1.5.0 · airockchip/rknn_model_zoo (github.com)
2、导出 ONNX 模型
# 切换到 PaddleDetection 或者 PaddleYOLO 源码目录下然后使用 tools/export_model.py 导出 paddle 模型
cd PaddleDetection
python tools/export_model.py -c configs/ppyoloe/ppyoloe_plus_crn_s_80e_coco.yml -o weightsppyoloe_plus_crn_s_80e_coco.pdparams exclude_nmsTrue exclude_post_processTrue --output_dir inference_model
# 其中 -c 是设置配置文件configs/目录下的配置文件
# --output_dir 指定模型保存目录默认是 output_inference
# -o 或者 --opt 设置配置选项这里设置了 weights 使用前面手动下载的权重等等
模型保存在 inference_model/ppyoloe_plus_crn_s_80e_coco 目录下
ppyoloe_plus_crn_s_80e_coco
├── infer_cfg.yml # 模型配置文件信息
├── model.pdiparams # 静态图模型参数
├── model.pdiparams.info # 参数额外信息一般无需关注
└── model.pdmodel # 静态图模型文件
然后将 paddle 模型转换成 ONNX 模型 pip install paddle2onnx
# 转换模型paddle2onnx --model_dir inference_model/ppyoloe_plus_crn_s_80e_coco --model_filename model.pdmodel --params_filename model.pdiparams --opset_version 11 --save_file ./inference_model/ppyoloe_plus_crn_s_80e_coco/ppyoloe_plus_crn_s_80e_coco.onnx
# 固定模型 shapepython -m paddle2onnx.optimize --input_model inference_model/ppyoloe_plus_crn_s_80e_coco/ppyoloe_plus_crn_s_80e_coco.onnx --output_model inference_model/ppyoloe_plus_crn_s_80e_coco/ppyoloe_plus_crn_s_80e_coco.onnx --input_shape_dict {image:[1,3,640,640]}
使用 Netron 查看下导出的ppyoloe_plus_crn_s_80e_coco.onnx模型的输出和输出 3、导出RKNN模型
使用 rknn-Toolkit2 工具将onnx 转换为 rknn模型并进行推理测试参考配套例程
代码参考rknn_model_zoo程序。
import os
import cv2
import sys
import numpy as np
from copy import copy
from rknn.api import RKNN
# model,image path
ONNX_MODEL ./ppyoloe_plus_crn_s_80e_coco.onnx
RKNN_MODEL ppyoloe_plus_crn_s_80e_coco.rknn
IMG_PATH ./test.jpg
#IMG_PATH ./000000087038.jpg
DATASET ./dataset.txt
QUANTIZE_ON True
OBJ_THRESH 0.5
NMS_THRESH 0.45
# OBJ_THRESH 0.001
# NMS_THRESH 0.65
IMG_SIZE (640, 640) # (width, height)
CLASSES (person, bicycle, car,motorbike ,aeroplane ,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,sofa,pottedplant,bed,diningtable,toilet ,tvmonitor,laptop ,mouse ,remote ,keyboard ,cell phone,microwave ,oven ,toaster,sink,refrigerator ,book,clock,vase,scissors ,teddy bear ,hair drier, toothbrush )
def filter_boxes(boxes, box_confidences, box_class_probs):Filter boxes with object threshold.box_confidences box_confidences.reshape(-1)
class_max_score np.max(box_class_probs, axis-1)classes np.argmax(box_class_probs, axis-1)
_class_pos np.where(class_max_score* box_confidences OBJ_THRESH)scores (class_max_score* box_confidences)[_class_pos]
boxes boxes[_class_pos]classes classes[_class_pos]
return boxes, classes, scores
def nms_boxes(boxes, scores):Suppress non-maximal boxes.# Returnskeep: ndarray, index of effective boxes.x boxes[:, 0]y boxes[:, 1]w boxes[:, 2] - boxes[:, 0]h boxes[:, 3] - boxes[:, 1]
areas w * horder scores.argsort()[::-1]
keep []while order.size 0:i order[0]keep.append(i)
xx1 np.maximum(x[i], x[order[1:]])yy1 np.maximum(y[i], y[order[1:]])xx2 np.minimum(x[i] w[i], x[order[1:]] w[order[1:]])yy2 np.minimum(y[i] h[i], y[order[1:]] h[order[1:]])
w1 np.maximum(0.0, xx2 - xx1 0.00001)h1 np.maximum(0.0, yy2 - yy1 0.00001)inter w1 * h1
ovr inter / (areas[i] areas[order[1:]] - inter)inds np.where(ovr NMS_THRESH)[0]order order[inds 1]keep np.array(keep)return keep
def dfl(position):# Distribution Focal Loss (DFL)import torchx torch.tensor(position)n,c,h,w x.shapep_num 4mc c//p_numy x.reshape(n,p_num,mc,h,w)y y.softmax(2)acc_metrix torch.tensor(range(mc)).float().reshape(1,1,mc,1,1)y (y*acc_metrix).sum(2)return y.numpy()
def box_process(position):grid_h, grid_w position.shape[2:4]col, row np.meshgrid(np.arange(0, grid_w), np.arange(0, grid_h))col col.reshape(1, 1, grid_h, grid_w)row row.reshape(1, 1, grid_h, grid_w)grid np.concatenate((col, row), axis1)stride np.array([IMG_SIZE[1]//grid_h, IMG_SIZE[0]//grid_w]).reshape(1,2,1,1)
position dfl(position)box_xy grid 0.5 -position[:,0:2,:,:]box_xy2 grid 0.5 position[:,2:4,:,:]xyxy np.concatenate((box_xy*stride, box_xy2*stride), axis1)
return xyxy
def post_process(input_data):boxes, scores, classes_conf [], [], []defualt_branch3pair_per_branch len(input_data)//defualt_branch# Python 忽略 score_sum 输出for i in range(defualt_branch):boxes.append(box_process(input_data[pair_per_branch*i]))classes_conf.append(input_data[pair_per_branch*i1])scores.append(np.ones_like(input_data[pair_per_branch*i1][:,:1,:,:], dtypenp.float32))
def sp_flatten(_in):ch _in.shape[1]_in _in.transpose(0,2,3,1)return _in.reshape(-1, ch)
boxes [sp_flatten(_v) for _v in boxes]classes_conf [sp_flatten(_v) for _v in classes_conf]scores [sp_flatten(_v) for _v in scores]
boxes np.concatenate(boxes)classes_conf np.concatenate(classes_conf)scores np.concatenate(scores)
# filter according to thresholdboxes, classes, scores filter_boxes(boxes, scores, classes_conf)
# nmsnboxes, nclasses, nscores [], [], []for c in set(classes):inds np.where(classes c)b boxes[inds]c classes[inds]s scores[inds]keep nms_boxes(b, s)
if len(keep) ! 0:nboxes.append(b[keep])nclasses.append(c[keep])nscores.append(s[keep])
if not nclasses and not nscores:return None, None, None
boxes np.concatenate(nboxes)classes np.concatenate(nclasses)scores np.concatenate(nscores)
return boxes, classes, scores
def draw(image, boxes, scores, classes):for box, score, cl in zip(boxes, scores, classes):top, left, right, bottom [int(_b) for _b in box]print(class: {}, score: {}.format(CLASSES[cl], score))print(box coordinate left,top,right,down: [{}, {}, {}, {}].format(top, left, right, bottom))
cv2.rectangle(image, (top, left), (right, bottom), (255, 0, 0), 2)cv2.putText(image, {:.2f}.format(score),(top, left - 6),cv2.FONT_HERSHEY_SIMPLEX,0.6, (0, 0, 255), 2)
def letter_box(im, new_shape, color(0, 0, 0)):# Resize and pad image while meeting stride-multiple constraintsshape im.shape[:2] # current shape [height, width]if isinstance(new_shape, int):new_shape (new_shape, new_shape)
# Scale ratio (new / old)r min(new_shape[0] / shape[0], new_shape[1] / shape[1])
# Compute paddingratio r # width, height ratiosnew_unpad int(round(shape[1] * r)), int(round(shape[0] * r))dw, dh new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1] # wh padding
dw / 2 # divide padding into 2 sidesdh / 2
if shape[::-1] ! new_unpad: # resizeim cv2.resize(im, new_unpad, interpolationcv2.INTER_LINEAR)top, bottom int(round(dh - 0.1)), int(round(dh 0.1))left, right int(round(dw - 0.1)), int(round(dw 0.1))im cv2.copyMakeBorder(im, top, bottom, left, right, cv2.BORDER_CONSTANT, valuecolor) # add borderreturn im, ratio, (dw, dh)
def get_real_box(src_shape, box, dw, dh, ratio):bbox copy(box)# unletter_box resultbbox[:,0] - dwbbox[:,0] / ratiobbox[:,0] np.clip(bbox[:,0], 0, src_shape[1])
bbox[:,1] - dhbbox[:,1] / ratiobbox[:,1] np.clip(bbox[:,1], 0, src_shape[0])
bbox[:,2] - dwbbox[:,2] / ratiobbox[:,2] np.clip(bbox[:,2], 0, src_shape[1])
bbox[:,3] - dhbbox[:,3] / ratiobbox[:,3] np.clip(bbox[:,3], 0, src_shape[0])return bbox
if __name__ __main__:# Create RKNN object#rknn RKNN(verboseTrue)rknn RKNN()
# pre-process config,target_platformrk3588print(-- Config model)rknn.config(mean_values[[0, 0, 0]], std_values[[255, 255, 255]], target_platformrk3588)print(done)
# Load ONNX modelprint(-- Loading model)ret rknn.load_onnx(modelONNX_MODEL)if ret ! 0:print(Load model failed!)exit(ret)print(done)
# Build modelprint(-- Building model)ret rknn.build(do_quantizationQUANTIZE_ON, datasetDATASET)if ret ! 0:print(Build model failed!)exit(ret)print(done)
# Export RKNN modelprint(-- Export rknn model)ret rknn.export_rknn(RKNN_MODEL)if ret ! 0:print(Export rknn model failed!)exit(ret)print(done)
# Init runtime environmentprint(-- Init runtime environment)ret rknn.init_runtime()# ret rknn.init_runtime(rk3566)if ret ! 0:print(Init runtime environment failed!)exit(ret)print(done)
# Set inputsimg_src cv2.imread(IMG_PATH)src_shape img_src.shape[:2]img, ratio, (dw, dh) letter_box(img_src, IMG_SIZE)img cv2.cvtColor(img, cv2.COLOR_BGR2RGB)#img cv2.resize(img_src, IMG_SIZE)
# Inferenceprint(-- Running model)outputs rknn.inference(inputs[img])print(done)
# post processboxes, classes, scores post_process(outputs)
img_p img_src.copy()if boxes is not None:draw(img_p, get_real_box(src_shape, boxes, dw, dh, ratio), scores, classes)cv2.imwrite(result.jpg, img_p)
五、部署测试
rknn_model_zoo测试前面有提及这里不在复现。自行测试。 六、参考连接
GitHub - PaddlePaddle/PaddleYOLO: YOLO series of PaddlePaddle implementation, PP-YOLOE, RT-DETR, YOLOv5, YOLOv6, YOLOv7, YOLOv8, YOLOX, YOLOv5u, YOLOv7u, YOLOv6Lite, RTMDet and so on.
https://github.com/airockchip/rknn_model_zoo/tree/main/models/CV/object_detection/yolo/
注意rknn_model_zoo使用的是PaddleDetection版本2.5这里使用的是PaddleYOLO版本2.6. 如有侵权或需要完整代码请及时联系博主。
