Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

pnnx converting model has permute layer missing #5679

Open
AnanasPizzaMigliore opened this issue Sep 3, 2024 · 0 comments
Open

pnnx converting model has permute layer missing #5679

AnanasPizzaMigliore opened this issue Sep 3, 2024 · 0 comments

Comments

@AnanasPizzaMigliore
Copy link

error log | 日志或报错信息 | ログ

model | 模型 | モデル

  1. original model

how to reproduce | 复现步骤 | 再現方法

1.last layer in ncnn param ConvolutionDepthWise convdwrelu_14 1 1 333 352 0=128 1=5 -23310=1,1.000000e-01 11=5 12=1 13=1 14=2 2=1 3=1 4=2 5=1 6=3200 7=128 9=2
Convolution convrelu_35 1 1 352 353 0=128 1=1 -23310=1,1.000000e-01 11=1 12=1 13=1 14=0 2=1 3=1 4=0 5=1 6=16384 9=2
ConvolutionDepthWise convdwrelu_15 1 1 353 354 0=128 1=5 -23310=1,1.000000e-01 11=5 12=1 13=1 14=2 2=1 3=1 4=2 5=1 6=3200 7=128 9=2
Convolution convrelu_36 1 1 354 355 0=128 1=1 -23310=1,1.000000e-01 11=1 12=1 13=1 14=0 2=1 3=1 4=0 5=1 6=16384 9=2
Convolution conv_94 1 1 355 356 0=33 1=1 11=1 12=1 13=1 14=0 2=1 3=1 4=0 5=1 6=4224
Reshape flatten_171 1 1 356 357 0=-1 1=33
Concat cat_12 4 1 339 345 351 357 out0 0=1
2.last layers in pnnx param nn.LeakyReLU head.cls_convs.3.1.act 1 1 274 275 negative_slope=1.000000e-01 #274=(1,128,20,13)f32 #275=(1,128,20,13)f32
nn.Conv2d convbn2d_113 1 1 275 276 bias=True dilation=(1,1) groups=1 in_channels=128 kernel_size=(1,1) out_channels=128 padding=(0,0) padding_mode=zeros stride=(1,1) @bias=(128)f32 @weight=(128,128,1,1)f32 $input=275 #275=(1,128,20,13)f32 #276=(1,128,20,13)f32
nn.LeakyReLU pnnx_unique_12 1 1 276 277 negative_slope=1.000000e-01 #276=(1,128,20,13)f32 #277=(1,128,20,13)f32
nn.Conv2d head.gfl_cls.3 1 1 277 278 bias=True dilation=(1,1) groups=1 in_channels=128 kernel_size=(1,1) out_channels=33 padding=(0,0) padding_mode=zeros stride=(1,1) @bias=(33)f32 @weight=(33,128,1,1)f32 #277=(1,128,20,13)f32 #278=(1,33,20,13)f32
torch.flatten torch.flatten_17 1 1 278 279 end_dim=-1 start_dim=2 $input=278 #278=(1,33,20,13)f32 #279=(1,33,260)f32
torch.cat torch.cat_13 4 1 249 259 269 279 280 dim=2 #249=(1,33,16000)f32 #259=(1,33,4000)f32 #269=(1,33,1000)f32 #279=(1,33,260)f32 #280=(1,33,21260)f32
Tensor.permute Tensor.permute_0 1 1 280 281 dims=(0,2,1) $input=280 #280=(1,33,21260)f32 #281=(1,21260,33)f32
pnnx.Output pnnx_output_0 1 0 281 #281=(1,21260,33)f32
3.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant