ThomasDelteil opened a new issue #10093: Cannot print summary of models loaded 
from onnx-mxnet
   ## Description
   Trying to print a model loaded from onnx-mxnet does not work because the 
value of `node["attrs"]["no_bias"]` is `"False"` rather than `False`.
   Not sure if an mxnet issue or onnx-mxnet issue
   ## Environment info (Required)
   MXNet 1.1.0
   Package used (Python/R/Scala/Julia):
   I am using python
   ## Minimum reproducible example
   Download the model here:
   sym, params = onnx_mxnet.import_model(onnx_path)
   ValueError                                Traceback (most recent call last)
   <ipython-input-253-99d037b1b5e0> in <module>()
   ----> 1 mx.visualization.print_summary(sym)
   ~/anaconda3/lib/python3.6/site-packages/mxnet/ in 
print_summary(symbol, shape, line_length, positions)
       182                 if key in shape_dict:
       183                     out_shape = shape_dict[key][1:]
   --> 184         total_params += print_layer_summary(nodes[i], out_shape)
       185         if i == len(nodes) - 1:
       186             print('=' * line_length)
   ~/anaconda3/lib/python3.6/site-packages/mxnet/ in 
print_layer_summary(node, out_shape)
       135         cur_param = 0
       136         if op == 'Convolution':
   --> 137             if ("no_bias" in node["attrs"]) and 
       138                 cur_param = pre_filter * 
       139                 for k in _str2tuple(node["attrs"]["kernel"]):
   ValueError: invalid literal for int() with base 10: 'False'

This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:

With regards,
Apache Git Services

Reply via email to