Hi3516(海思)训练yolov5-6.0--＞转oonx--＞转caffe-转.wk文件_yolov5 海思3516

1 训练

1 更改模型结构，将Upsample变成ConvTranspose2d,

# YOLOv5 v6.0 head
head:
  [[-1, 1, Conv, [512, 1, 1]],
#   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [-1, -1, nn.ConvTranspose2d,[256, 256, 2, 2]],
   [[-1, 6], 1, Concat, [1]],  # cat backbone P4
   [-1, 3, C3, [512, False]],  # 13
 
   [-1, 1, Conv, [256, 1, 1]],
#   [-1, 1, nn.Upsample, [None, 2, 'nearest']],
   [-1, -1, nn.ConvTranspose2d,[128, 128, 2, 2]],
   [[-1, 4], 1, Concat, [1]],  # cat backbone P3
   [-1, 3, C3, [256, False]],  # 17 (P3/8-small)

二. 导出模型

1. 导出onnx模型：
(1) 在export中opset改为9
(2) 在models/yolo.py中修改detect中代码如下：

# 为海思3561dv300更改 train use 官方
class Detect(nn.Module):
    stride = None  # strides computed during build
    onnx_dynamic = False  # ONNX export parameter
 
    def __init__(self, nc=80, anchors=(), ch=(), inplace=True):  # detection layer
        super().__init__()
        self.nc = nc  # number of classes
        self.no = nc + 5  # number of outputs per anchor
        self.nl = len(anchors)  # number of detection layers
        self.na = len(anchors[0]) // 2  # number of anchors
        self.grid = [torch.zeros(1)] * self.nl  # init grid
        self.anchor_grid = [torch.zeros(1)] * self.nl  # init anchor grid
        self.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2))  # shape(nl,na,2)
        self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch)  # output conv
        self.inplace = inplace  # use in-place ops (e.g. slice assignment)
 
    def forward(self, x):
        # print_feature=2
        z = []  # inference output
        for i in range(self.nl):
            x[i] = self.m[i](x[i])  # conv
            bs, _, ny, nx = x[i].shape  # x(bs,255,20,20) to x(bs,3,20,20,85)
           # x[i] = self.m[i](x[i])  # **增加这行代码**
           #  x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
            x[i] = x[i].view(bs, self.na, self.no, ny*nx)
 
            if not self.training:  # inference
                if self.grid[i].shape[2:4] != x[i].shape[2:4] or self.onnx_dynamic:
                    self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i)
 
                # y = x[i].sigmoid()
                y = x[i]
                # if self.inplace:
                #     y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i]  # xy
                #     y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i]  # wh
                # else:  # for YOLOv5 on AWS Inferentia https://github.com/ultralytics/yolov5/pull/2953
                #     xy = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i]  # xy
                #     wh = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i]  # wh
                #     y = torch.cat((xy, wh, y[..., 4:]), -1)
                # z.append(y.view(bs, -1, self.no))
                z.append(y)
        return z
        # return x if self.training else (torch.cat(z, 1), x)
 
    def _make_grid(self, nx=20, ny=20, i=0):
        d = self.anchors[i].device
        yv, xv = torch.meshgrid([torch.arange(ny).to(d), torch.arange(nx).to(d)])
        grid = torch.stack((xv, yv), 2).expand((1, self.na, ny, nx, 2)).float()
        anchor_grid = (self.anchors[i].clone() * self.stride[i]) \
            .view((1, self.na, 1, 1, 2)).expand((1, self.na, ny, nx, 2)).float()
        return grid, anchor_grid
 
# 官方原先的DETAECT
class Detect1(nn.Module):
    stride = None  # strides computed during build
    onnx_dynamic = False  # ONNX export parameter
 
    def __init__(self, nc=80, anchors=(), ch=(), inplace=True):  # detection layer
        super().__init__()
        self.nc = nc  # number of classes
        self.no = nc + 5  # number of outputs per anchor
        self.nl = len(anchors)  # number of detection layers
        self.na = len(anchors[0]) // 2  # number of anchors
        self.grid = [torch.zeros(1)] * self.nl  # init grid
        self.anchor_grid = [torch.zeros(1)] * self.nl  # init anchor grid
        self.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2))  # shape(nl,na,2)
        self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch)  # output conv
        self.inplace = inplace  # use in-place ops (e.g. slice assignment)
 
    def forward(self, x):
        z = []  # inference output
        for i in range(self.nl):
            x[i] = self.m[i](x[i])  # conv
            bs, _, ny, nx = x[i].shape  # x(bs,255,20,20) to x(bs,3,20,20,85)
            x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()
 
            if not self.training:  # inference
                if self.grid[i].shape[2:4] != x[i].shape[2:4] or self.onnx_dynamic:
                    self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i)
 
                y = x[i].sigmoid()
                if self.inplace:
                    y[..., 0:2] = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i]  # xy
                    y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i]  # wh
                else:  # for YOLOv5 on AWS Inferentia https://github.com/ultralytics/yolov5/pull/2953
                    xy = (y[..., 0:2] * 2. - 0.5 + self.grid[i]) * self.stride[i]  # xy
                    wh = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i]  # wh
                    y = torch.cat((xy, wh, y[..., 4:]), -1)
                z.append(y.view(bs, -1, self.no))
 
        return x if self.training else (torch.cat(z, 1), x)
 
    def _make_grid(self, nx=20, ny=20, i=0):
        d = self.anchors[i].device
        yv, xv = torch.meshgrid([torch.arange(ny).to(d), torch.arange(nx).to(d)])
        grid = torch.stack((xv, yv), 2).expand((1, self.na, ny, nx, 2)).float()
        anchor_grid = (self.anchors[i].clone() * self.stride[i]) \
            .view((1, self.na, 1, 1, 2)).expand((1, self.na, ny, nx, 2)).float()
        return grid, anchor_grid

改动有以下几点：
1：去掉了原先的permute；
2：将view原来的输出维度(bs, na, no, ny, nx) 改为 （bs, na, no, ny * nx）；
3：去除了后处理坐标点和宽高decode代码，去除cat操作

现在来分析下为什么这么改:
1：nnie不支持5个维度的permute（即transpose），且只支持0231的方式，过于局限，我们不妨删掉这一层，在后处理中按照合适的读取方式去找结果就好了。
2：nnie的reshape也只支持4维，且第一维必须是0，为了能用nnie的reshape，我们不得不把x和y共享一个维度，这导致的结果是输出结果中，x和y在同一行，我们只需按个数取值即可。
3：后处理中，对三个检测层分别处理，所以不需要concat

执行

python export.py --opset 9 --imgsz 640  640  --simplify --weights best.pt

导出onnx-sim模型

python -m onnxsim xxx.onnx xxx-sim.onnx

3 转caffe

github上搜索 yolov5_onnx2caffe 项目（https://github.com/Wulingtian/yolov5_onnx2caffe）

vim convertCaffe.py

设置onnx_path（上面转换得到的onnx模型），prototxt_path（caffe的prototxt保存路径），caffemodel_path（caffe的caffemodel保存路径）

执行

python ./yolov5_onnx2caffe/convertCaffe.py

4 caffe转 wk文件

 2

 注意将.cfg文件中compile_mode = 0 改成compile_mode = 1

 说明：

image_list 字段表示测试的数据，具体如下

 

 点击运行