Carafe

[Luckycat518]

❔Question

你好，我在尝试采用carafe算子进行训练时，报错如下: yolo.py in parsemodel c2 = make_divisible(c2 gw, 8) TypeError: unsupported operand type(s) for : 'NoneType' and 'float' 请问该如何解决呢？是哪里没设置对吗?

Additional context

[positive666]

❔Question

你好，我在尝试采用carafe算子进行训练时，报错如下: yolo.py in parsemodel c2 = make_divisible(c2 gw, 8) TypeError: unsupported operand type(s) for : 'NoneType' and 'float' 请问该如何解决呢？是哪里没设置对吗?

Additional context

你发下命令 看看

[positive666]

❔Question

你好，我在尝试采用carafe算子进行训练时，报错如下: yolo.py in parsemodel c2 = make_divisible(c2 gw, 8) TypeError: unsupported operand type(s) for : 'NoneType' and 'float' 请问该如何解决呢？是哪里没设置对吗?

Additional context

你发下命令 看看

也可以先clone我新的代码 因为昨天进行了一次大更新

[Luckycat518]

你好，我是基于您的代码对我的v6.0的common.py和yolo.py进行了修改。具体修改如下

在common.py中添加了如下内容： class CARAFE(nn.Module):

CARAFE: Content-Aware ReAssembly of FEatures https://arxiv.org/pdf/1905.02188.pdf

def __init__(self, c1, c2, kernel_size=3, up_factor=2):
    super(CARAFE, self).__init__()
    self.kernel_size = kernel_size
    self.up_factor = up_factor
    self.down = nn.Conv2d(c1, c1 // 4, 1)
    self.encoder = nn.Conv2d(c1 // 4, self.up_factor ** 2 * self.kernel_size ** 2,
                             self.kernel_size, 1, self.kernel_size // 2)
    self.out = nn.Conv2d(c1, c2, 1)

def forward(self, x):
    N, C, H, W = x.size()
    # N,C,H,W -> N,C,delta*H,delta*W
    # kernel prediction module
    kernel_tensor = self.down(x)  # (N, Cm, H, W)
    kernel_tensor = self.encoder(kernel_tensor)  # (N, S^2 * Kup^2, H, W)
    kernel_tensor = F.pixel_shuffle(kernel_tensor, self.up_factor)  # (N, S^2 * Kup^2, H, W)->(N, Kup^2, S*H, S*W)
    kernel_tensor = F.softmax(kernel_tensor, dim=1)  # (N, Kup^2, S*H, S*W)
    kernel_tensor = kernel_tensor.unfold(2, self.up_factor, step=self.up_factor) # (N, Kup^2, H, W*S, S)
    kernel_tensor = kernel_tensor.unfold(3, self.up_factor, step=self.up_factor) # (N, Kup^2, H, W, S, S)
    kernel_tensor = kernel_tensor.reshape(N, self.kernel_size ** 2, H, W, self.up_factor ** 2) # (N, Kup^2, H, W, S^2)
    kernel_tensor = kernel_tensor.permute(0, 2, 3, 1, 4)  # (N, H, W, Kup^2, S^2)

    # content-aware reassembly module
    # tensor.unfold: dim, size, step
    x = F.pad(x, pad=(self.kernel_size // 2, self.kernel_size // 2,
                                      self.kernel_size // 2, self.kernel_size // 2),
                      mode='constant', value=0) # (N, C, H+Kup//2+Kup//2, W+Kup//2+Kup//2)
    x = x.unfold(2, self.kernel_size, step=1) # (N, C, H, W+Kup//2+Kup//2, Kup)
    x = x.unfold(3, self.kernel_size, step=1) # (N, C, H, W, Kup, Kup)
    x = x.reshape(N, C, H, W, -1) # (N, C, H, W, Kup^2)
    x = x.permute(0, 2, 3, 1, 4)  # (N, H, W, C, Kup^2)

    out_tensor = torch.matmul(x, kernel_tensor)  # (N, H, W, C, S^2)
    out_tensor = out_tensor.reshape(N, H, W, -1)
    out_tensor = out_tensor.permute(0, 3, 1, 2)
    out_tensor = F.pixel_shuffle(out_tensor, self.up_factor)
    out_tensor = self.out(out_tensor)
    #print("up shape:",out_tensor.shape)
    return out_tensor

在yolo.py中添加了CARAFE，修改如下： if m in [Conv, GhostConv, Bottleneck, GhostBottleneck, SPP, SPPF, GSConv, DWConv, CARAFE, MixConv2d, Focus, CrossConv, BottleneckCSP, C3, C3TR, C3STR, C3SPP, C3Ghost, CBAM, Conv_maxpool, ShuffleNetV2_InvertedResidual......

运行的具体指令如下： python train.py --weights weights/yolov5l.pt --cfg models/ablation_experimental_wgq/yolov5lCBAM-SwinTrans-DWconv-Decoupledhead-CARAFE.yaml --data data/Ws.yaml --epoch 300 --batch-size 8 --img 608 --nohalf --device '0'

[Luckycat518]

grey_questionQuestion

你好，我在尝试采用carafe算子进行训练时，报错如下: yolo.py in parsemodel c2 = make_divisible(c2 gw, 8) TypeError: unsupported operand type(s) for : 'NoneType' and 'float' 请问该如何解决呢？是哪里没设置对吗?

Additional context

你发下命令 看看

也可以先clone我新的代码 因为昨天进行了一次大更新

好的，我看看，您有空也可以看看我结合您代码进行的更改是否存在问题。即使不结合trans，仅在yolov5-6.0.yaml中基于上述修改替换CARAFE算子依然报上述错误“: yolo.py in parsemodel c2 = make_divisible(c2 gw, 8) TypeError: unsupported operand type(s) for : 'NoneType' and 'float' ”

[Luckycat518]

grey_questionQuestion

你好，我在尝试采用carafe算子进行训练时，报错如下: yolo.py in parsemodel c2 = make_divisible(c2 gw, 8) TypeError: unsupported operand type(s) for : 'NoneType' and 'float' 请问该如何解决呢？是哪里没设置对吗?

Additional context

你发下命令 看看

也可以先clone我新的代码 因为昨天进行了一次大更新

clone您新的代码并在您的文件夹下运行指令“python train.py --weights weights/yolov5s.pt --cfg models/yolov5s-carafe.yaml --data data/Weld2009.yaml --epoch 300 --batch-size 8 --img 608 --device '0' --name l-Cara”后报错如下“File "train.py", line 416, in train loss, loss_items = compute_loss(pred, targets.to(device)) # loss scaled by batch_size File "/home/imcm/code/wanggq/yolov5research-master/utils/loss.py", line 221, in call pxy, pwh, , pcls = pi[b, a, gj, gi].tensor_split((2, 4, 5), dim=1) # target-subset of predictions AttributeError: 'Tensor' object has no attribute 'tensor_split' ”

[positive666]

grey_questionQuestion

你好，我在尝试采用carafe算子进行训练时，报错如下: yolo.py in parsemodel c2 = make_divisible(c2 gw, 8) TypeError: unsupported operand type(s) for : 'NoneType' and 'float' 请问该如何解决呢？是哪里没设置对吗?

Additional context

你发下命令 看看

也可以先clone我新的代码 因为昨天进行了一次大更新

clone您新的代码并在您的文件夹下运行指令“python train.py --weights weights/yolov5s.pt --cfg models/yolov5s-carafe.yaml --data data/Weld2009.yaml --epoch 300 --batch-size 8 --img 608 --device '0' --name l-Cara”后报错如下“File "train.py", line 416, in train loss, loss_items = compute_loss(pred, targets.to(device)) # loss scaled by batch_size File "/home/imcm/code/wanggq/yolov5research-master/utils/loss.py", line 221, in call pxy, pwh, , pcls = pi[b, a, gj, gi].tensor_split((2, 4, 5), dim=1) # target-subset of predictions AttributeError: 'Tensor' object has no attribute 'tensor_split' ”

好的 今天更新 不过我先测下yolov7的结构

[github-actions[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.