Yolov5 转换成 RKNN模型_yolov5转rknn

1.前言

关于模型转换后在NPU上运行，看了很多的教程，但是对于不熟悉模型转换的小白，在转换的几个关键点还是要注意的，所以本次的教程从最基本的开始做！！！

2.环境准备

(1) RKNN的环境配置好(推荐使用1.7.1版本，环境都没有配置好，那可能要努力一下了，都有教程的)

(2)下载yolov5代码yolov5官方链接，相关的模型下载如图所示,本次教程下载了5s/5m/5x的模型

          python环境的配置，推荐使用torch==1.8、torchvision==0.9.1、其他的环境

3.RKNN模型转换

     

 3.1 将yolov5s.pt、yolov5m.pt、yolov5x.pt拷贝到下yolov5代码目录

  (1)执行   python3 export.py --weights yolov5s.pt --img 640 --batch 1 --opset 12
  生成 yolov5s.onnx模型，换成5m或者5x就把名字换一下，相信大家都会哒！

  (2)运行的结果如下图所示:

         好啦，我们已经完成了onnx的模型导出，已经完成50%了！！！

         注：yolov5工程需要使用pytorch 1.8.0 或 1.9.0 版本才能正常导出。

3.2将onnx模型准换成rknn

        来，上代码！！！下面的一些参数大家还是懂的哈，不懂的话，那再去看看rknn的文档(相信你们都到这一步了，都懂的)

ONNX_MODEL = 'yolov5s.onnx'
RKNN_MODEL = 'yolov5s.rknn'
IMG_PATH = './bus.jpg'
DATASET = './dataset.txt'

重点来了，绝对的干货！！！换成5m或者5x后，有几个节点需要修改的

(1)yolov5s导出RKNN的节点

        ret = rknn.load_onnx(model=ONNX_MODEL,outputs=['396', '440', '484'])

(2)yolov5m导出RKNN的节点

        ret = rknn.load_onnx(model=ONNX_MODEL,outputs=['462', '506', '550'])

(3)yolov5x导出RKNN的节点

        ret = rknn.load_onnx(model=ONNX_MODEL,outputs=['696', '740', '784'])

注意: 1. 这3个节点的输出shape分别是[1, 3, 80, 80, 85],[1, 3, 40, 40, 85],[1, 3, 20, 20, 85] 

           2.具体的网络结构可以通过netron查看(哎,我还是截图出来看看吧，谁叫这是保姆教程呢)

          Example: 以yolov5s为例，看到里面的transpose算子了吧，是不是有三个！然后看outputs的name是396,再看其他两个transpose的name分别是440和484，好啦，已经很清楚了，在不知道就枪毙掉！！！

         ONNX转rknn代码(yolov5s)，转5m和5x只需要修改ONNX_MODEL、RKNN_MODEL、导出的节点就行了

        还有一个坑，就是在转化的时候保证onnx==1.6.0！！！

import numpy as np
import cv2
from rknn.api import RKNN
import os
 
 
ONNX_MODEL = 'yolov5s.onnx'
RKNN_MODEL = 'yolov5s.rknn'
IMG_PATH = './bus.jpg'
DATASET = './dataset.txt'
 
 
 
if __name__ == '__main__':
 
    # Create RKNN object
    rknn = RKNN()
 
    if not os.path.exists(ONNX_MODEL):
        print('model not exist')
        exit(-1)
    
    # pre-process config
    print('--> Config model')
    rknn.config(reorder_channel='0 1 2',mean_values=[[0, 0, 0]],std_values=[[255, 255, 255]],optimization_level=3,target_platform = 'rk1808',output_optimize=1)
    print('done')
 
    # Load ONNX model
    print('--> Loading model')
    ret = rknn.load_onnx(model=ONNX_MODEL,outputs=['396', '440', '484'])
    if ret != 0:
        print('Load yolov5 failed!')
        exit(ret)
    print('done')
 
    # Build model
    print('--> Building model')
    ret = rknn.build(do_quantization=True, dataset=DATASET)
    if ret != 0:
        print('Build yolov5 failed!')
        exit(ret)
    print('done')
 
    # Export RKNN model
    print('--> Export RKNN model')
    ret = rknn.export_rknn(RKNN_MODEL)
    if ret != 0:
        print('Export yolov5rknn failed!')
        exit(ret)
    print('done')

        下图是转好的模型(嘿嘿，还是有亿点点成就的)

 4.rknn模型推理

        准备好图片、模型、推理代码就开始了，是不是很简单(感觉rknn的工具已经很傻瓜化了，就是环境要配置好)

import os
import urllib
import traceback
import time
import sys
import numpy as np
import cv2
from rknn.api import RKNN
 
 
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 sigmoid(x):
    return 1 / (1 + np.exp(-x))
 
def xywh2xyxy(x):
    # Convert [x, y, w, h] to [x1, y1, x2, y2]
    y = np.copy(x)
    y[:, 0] = x[:, 0] - x[:, 2] / 2  # top left x
    y[:, 1] = x[:, 1] - x[:, 3] / 2  # top left y
    y[:, 2] = x[:, 0] + x[:, 2] / 2  # bottom right x
    y[:, 3] = x[:, 1] + x[:, 3] / 2  # bottom right y
    return y
 
def process(input, mask, anchors):
 
    anchors = [anchors[i] for i in mask]
    grid_h, grid_w = map(int, input.shape[0:2])
 
    box_confidence = sigmoid(input[..., 4])
    box_confidence = np.expand_dims(box_confidence, axis=-1)
 
    box_class_probs = sigmoid(input[..., 5:])
 
    box_xy = sigmoid(input[..., :2])*2 - 0.5
 
    col = np.tile(np.arange(0, grid_w), grid_w).reshape(-1, grid_w)
    row = np.tile(np.arange(0, grid_h).reshape(-1, 1), grid_h)
    col = col.reshape(grid_h, grid_w, 1, 1).repeat(3, axis=-2)
    row = row.reshape(grid_h, grid_w, 1, 1).repeat(3, axis=-2)
    grid = np.concatenate((col, row), axis=-1)
    box_xy += grid
    box_xy *= int(IMG_SIZE/grid_h)
 
    box_wh = pow(sigmoid(input[..., 2:4])*2, 2)
    box_wh = box_wh * anchors
 
    box = np.concatenate((box_xy, box_wh), axis=-1)
 
    return box, box_confidence, box_class_probs
 
def filter_boxes(boxes, box_confidences, box_class_probs):
    """Filter boxes with box threshold. It's a bit different with origin yolov5 post process!
    # Arguments
        boxes: ndarray, boxes of objects.
        box_confidences: ndarray, confidences of objects.
        box_class_probs: ndarray, class_probs of objects.
    # Returns
        boxes: ndarray, filtered boxes.
        classes: ndarray, classes for boxes.
        scores: ndarray, scores for boxes.
    """
    box_classes = np.argmax(box_class_probs, axis=-1)
    box_class_scores = np.max(box_class_probs, axis=-1)
    pos = np.where(box_confidences[...,0] >= BOX_THRESH)
 
 
    boxes = boxes[pos]
    classes = box_classes[pos]
    scores = box_class_scores[pos]
 
    return boxes, classes, scores
 
def nms_boxes(boxes, scores):
    """Suppress non-maximal boxes.
    # Arguments
        boxes: ndarray, boxes of objects.
        scores: ndarray, scores of objects.
    # Returns
        keep: ndarray, index of effective boxes.
    """
    x = boxes[:, 0]
    y = boxes[:, 1]
    w = boxes[:, 2] - boxes[:, 0]
    h = boxes[:, 3] - boxes[:, 1]
 
    areas = w * h
    order = 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 yolov5_post_process(input_data):
    masks = [[0, 1, 2], [3, 4, 5], [6, 7, 8]]
    anchors = [[10, 13], [16, 30], [33, 23], [30, 61], [62, 45],
              [59, 119], [116, 90], [156, 198], [373, 326]]
 
    boxes, classes, scores = [], [], []
    for input,mask in zip(input_data, masks):
        b, c, s = process(input, mask, anchors)
        b, c, s = filter_boxes(b, c, s)
        boxes.append(b)
        classes.append(c)
        scores.append(s)
 
    boxes = np.concatenate(boxes)
    boxes = xywh2xyxy(boxes)
    classes = np.concatenate(classes)
    scores = np.concatenate(scores)
 
    nboxes, 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)
 
        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):
    """Draw the boxes on the image.
    # Argument:
        image: original image.
        boxes: ndarray, boxes of objects.
        classes: ndarray, classes of objects.
        scores: ndarray, scores of objects.
        all_classes: all classes name.
    """
    for box, score, cl in zip(boxes, scores, classes):
        top, left, right, bottom = box
        # print('class: {}, score: {}'.format(CLASSES[cl], score))
        # print('box coordinate left,top,right,down: [{}, {}, {}, {}]'.format(top, left, right, bottom))
        top = int(top)
        left = int(left)
        right = int(right)
        bottom = int(bottom)
 
        cv2.rectangle(image, (top, left), (right, bottom), (255, 0, 0), 2)
        cv2.putText(image, '{0} {1:.2f}'.format(CLASSES[cl], score),
                    (top, left - 6),
                    cv2.FONT_HERSHEY_SIMPLEX,
                    0.6, (0, 0, 255), 2)
 
 
def letterbox(im, new_shape=(640, 640), color=(0, 0, 0)):
    # Resize and pad image while meeting stride-multiple constraints
    shape = 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 padding
    ratio = r, r  # width, height ratios
    new_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 sides
    dh /= 2
 
    if shape[::-1] != new_unpad:  # resize
        im = cv2.resize(im, new_unpad, interpolation=cv2.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, value=color)  # add border
    return im, ratio, (dw, dh)
 
 
if __name__ == '__main__':
 
    RKNN_MODEL = 'yolov5s.rknn'
    IMG_PATH = './bus.jpg'
    BOX_THRESH = 0.5
    NMS_THRESH = 0.45
    IMG_SIZE = 640
 
    # Create RKNN object
    rknn = RKNN()
    ret = rknn.load_rknn(RKNN_MODEL)
    # init runtime environment
    print('--> Init runtime environment')
    ret = rknn.init_runtime(target='rk1808')
    if ret != 0:
        print('Init runtime environment failed')
        exit(ret)
    print('done')
 
    # Set inputs
    img = cv2.imread(IMG_PATH)
    img, ratio, (dw, dh) = letterbox(img, new_shape=(IMG_SIZE, IMG_SIZE))
    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    img = cv2.resize(img,(IMG_SIZE, IMG_SIZE))
 
    # Inference
    print('--> Running model')
    t1 = time.time()
    outputs = rknn.inference(inputs=[img])
    t2 = time.time()
    print('DET_inf_time:', t2 - t1)
 
    # post process
    input0_data = outputs[0]
    input1_data = outputs[1]
    input2_data = outputs[2]
 
 
    input0_data = input0_data.reshape([3, 80, 80, 85])
    input1_data = input1_data.reshape([3, 40, 40, 85])
    input2_data = input2_data.reshape([3, 20, 20, 85])
 
    input_data = list()
    input_data.append(np.transpose(input0_data, (1, 2, 0, 3)))
    input_data.append(np.transpose(input1_data, (1, 2, 0, 3)))
    input_data.append(np.transpose(input2_data, (1, 2, 0, 3)))
 
    boxes, classes, scores = yolov5_post_process(input_data)
    t3 = time.time()
    print('post_process_time:',t3 - t2)
    img_1 = cv2.cvtColor(img, cv2.COLOR_RGB2BGR)
    if boxes is not None:
        draw(img_1, boxes, scores, classes)
    # cv2.imshow("post process result", img_1)
    # cv2.waitKeyEx(0)
    cv2.imwrite('result.jpg', img_1)
    rknn.release()

推理结果如下，结果还是挺不错的