rknn1.4.0转pytorch模型报除0错误

模型是deep_sort_pytorch   https://github.com/ZQPei/deep_sort_pytorch

保存成torchscript模型：

1.     net = Net(reid=True)
   
2.     model_path = "/home/fly/ai/project/led/deep_sort_pytorch-master/deep_sort/deep/checkpoint/ckpt.t7"
   
3.     state_dict = torch.load(model_path, map_location=lambda storage, loc: storage)['net_dict']
   
4.     net.load_state_dict(state_dict)
   
5.     x = torch.randn(1,3,128,64)
   
6.     torch.jit.trace(net, x).save("./deep_sort.pt")

复制代码
转换代码

1. input_size_list = [[3, 128, 64]]
   
2. 
   
3.     # Create RKNN object
   
4.     rknn = RKNN()
   
5. 
   
6.     # pre-process config
   
7.     print('--> config model')
   
8.     rknn.config(reorder_channel='0 1 2')
   
9.     print('done')
   
10. 
    
11.     # Load pytorch model
    
12.     print('--> Loading model')
    
13.     ret = rknn.load_pytorch(model=model, input_size_list=input_size_list)

复制代码

错误日志

1. <p>
   
2. </p><p>/home/fly/ai/project/led/deep_sort_pytorch-master/deep_sort/deep/deep_sort.pt ********************</p><p>WARNING: Token 'COMMENT' defined, but not used</p><p>WARNING: There is 1 unused token</p><p>Load pytorch model failed!</p><p>E Catch exception when loading pytorch model: /home/fly/ai/project/led/deep_sort_pytorch-master/deep_sort/deep/deep_sort.pt!</p><p>E Traceback (most recent call last):</p><p>E   File "rknn/api/rknn_base.py", line 657, in rknn.api.rknn_base.RKNNBase.load_pytorch</p><p>E   File "rknn/base/RKNNlib/app/importer/import_pytorch.py", line 97, in rknn.base.RKNNlib.app.importer.import_pytorch.ImportPytorch.run</p><p>E   File "rknn/base/RKNNlib/converter/convert_pytorch.py", line 755, in rknn.base.RKNNlib.converter.convert_pytorch.convert_pytorch.__init__</p><p>E   File "rknn/base/RKNNlib/converter/convert_pytorch.py", line 839, in rknn.base.RKNNlib.converter.convert_pytorch.convert_pytorch.model_simplify</p><p>E   File "rknn/base/RKNNlib/converter/convert_pytorch.py", line 124, in rknn.base.RKNNlib.converter.convert_pytorch.torch_inference_engine.shape_pick</p><p>E   File "rknn/base/RKNNlib/converter/convert_pytorch.py", line 159, in rknn.base.RKNNlib.converter.convert_pytorch.torch_inference_engine.__ir_shape_inference</p><p>E   File "rknn/base/RKNNlib/converter/convert_pytorch.py", line 233, in rknn.base.RKNNlib.converter.convert_pytorch.torch_inference_engine.elementwise_boardcast_shape</p><p>E   File "rknn/base/RKNNlib/converter/convert_pytorch.py", line 124, in rknn.base.RKNNlib.converter.convert_pytorch.torch_inference_engine.shape_pick</p><p>E   File "rknn/base/RKNNlib/converter/convert_pytorch.py", line 159, in rknn.base.RKNNlib.converter.convert_pytorch.torch_inference_engine.__ir_shape_inference</p><p>E   File "rknn/base/RKNNlib/converter/convert_pytorch.py", line 281, in rknn.base.RKNNlib.converter.convert_pytorch.torch_inference_engine.view_shape</p><p>E ZeroDivisionError: integer division or modulo by zero</p>

复制代码
python环境

1. torch               1.2.0
   
2. torchvision         0.4.0
   
3. rknn-toolkit        1.4.0

复制代码请问这个除零错误是什么原因呢？

把div的后处理从网络中剥离掉，就可以了。