Core dumped

[Zrufy]

This is the model converted from pytorch to onnx to ncnn.

7767517 36 36 Input input.1 0 1 input.1 Convolution 61 1 1 input.1 61 0=24 1=5 11=5 2=1 12=1 3=2 13=2 4=2 14=2 15=2 16=2 5=1 6=600 ReLU 62 1 1 61 62 ConvolutionDepthWise 63 1 1 62 63 0=24 1=3 11=3 2=1 12=1 3=1 13=1 4=1 14=1 15=1 16=1 5=1 6=216 7=24 ReLU 64 1 1 63 64 Convolution 65 1 1 64 65 0=128 1=1 11=1 2=1 12=1 3=1 13=1 4=0 14=0 15=0 16=0 5=1 6=3072 Pooling 66 1 1 65 66 0=0 1=2 11=2 2=2 12=2 3=0 13=0 14=0 15=0 5=1 ConvolutionDepthWise 67 1 1 66 67 0=128 1=3 11=3 2=1 12=1 3=1 13=1 4=1 14=1 15=1 16=1 5=1 6=1152 7=128 BatchNorm 68 1 1 67 68 0=128 ReLU 69 1 1 68 69 Convolution 70 1 1 69 70 0=256 1=1 11=1 2=1 12=1 3=1 13=1 4=0 14=0 15=0 16=0 5=1 6=32768 BatchNorm 71 1 1 70 71 0=256 ConvolutionDepthWise 72 1 1 71 72 0=256 1=3 11=3 2=1 12=1 3=1 13=1 4=1 14=1 15=1 16=1 5=1 6=2304 7=256 ReLU 73 1 1 72 73 Convolution 74 1 1 73 74 0=256 1=1 11=1 2=1 12=1 3=1 13=1 4=0 14=0 15=0 16=0 5=1 6=65536 Pooling 75 1 1 74 75 0=0 1=2 11=2 2=1 12=2 3=1 13=0 14=1 15=0 5=1 ConvolutionDepthWise 76 1 1 75 76 0=256 1=3 11=3 2=1 12=1 3=1 13=1 4=1 14=1 15=1 16=1 5=1 6=2304 7=256 BatchNorm 77 1 1 76 77 0=256 ReLU 78 1 1 77 78 Convolution 79 1 1 78 79 0=512 1=1 11=1 2=1 12=1 3=1 13=1 4=0 14=0 15=0 16=0 5=1 6=131072 BatchNorm 80 1 1 79 80 0=512 ConvolutionDepthWise 81 1 1 80 81 0=512 1=3 11=3 2=1 12=1 3=1 13=1 4=1 14=1 15=1 16=1 5=1 6=4608 7=512 ReLU 82 1 1 81 82 Convolution 83 1 1 82 83 0=512 1=1 11=1 2=1 12=1 3=1 13=1 4=0 14=0 15=0 16=0 5=1 6=262144 Pooling 84 1 1 83 84 0=0 1=2 11=2 2=1 12=2 3=1 13=0 14=1 15=0 5=1 ConvolutionDepthWise 85 1 1 84 85 0=512 1=2 11=2 2=1 12=1 3=1 13=1 4=0 14=0 15=0 16=0 5=1 6=2048 7=512 BatchNorm 86 1 1 85 86 0=512 ReLU 87 1 1 86 87 Convolution 88 1 1 87 88 0=512 1=1 11=1 2=1 12=1 3=1 13=1 4=0 14=0 15=0 16=0 5=1 6=262144 BatchNorm 89 1 1 88 89 0=512 Squeeze 90 1 1 89 90 -23303=1,2 Permute 91 1 1 90 91 Reshape 105 1 1 91 105 0=512 InnerProduct 106 1 1 105 106 0=128 1=1 2=65536 InnerProduct 107 1 1 106 107 0=37 1=1 2=4736 Reshape 113 1 1 107 113 0=-1 1=1

when i run the model this go to core dumped.Any help to figure out? i use the same model dense used by https://github.com/ouyanghuiyu/chineseocr_lite. @nihui

[Zrufy]

@nihui can you help me?

[nihui]

针对onnx模型转换的各种问题，推荐使用最新的pnnx工具转换到ncnn In view of various problems in onnx model conversion, it is recommended to use the latest pnnx tool to convert your model to ncnn

pip install pnnx
pnnx model.onnx inputshape=[1,3,224,224]

详细参考文档 Detailed reference documentation https://github.com/pnnx/pnnx https://github.com/Tencent/ncnn/wiki/use-ncnn-with-pytorch-or-onnx#how-to-use-pnnx