Zhibo Liu

Same error, try to change to g++-4.8 (maybe also gcc-4.8), it may help.

If I follow the steps described in the [video](https://www.youtube.com/watch?v=BpFs83MU3HM&feature=youtu.be) you provided and compile the onnx model under the [directory](https://github.com/ai-techsystems/deepC/tree/master/test/compiler/mnist) with command `compile-onnx`, it will generate a cpp file and *.weight...

Thank you @carlopi for answering. The asmjs code compiled with "-g" option is still hard to understand. Also, we prefer to compile wasm code for better performance. So sorry, this...

Thank you very much for your help. We only use `-O0` option to investigate the availability of dwarf debug information. We will consider turning to asmjs as well. Thanks!

Thank you for your suggestion! We are consulting with the university and will add a license to BTD shortly.

A license is created now. Please see [LICENSE](https://github.com/monkbai/DNN-decompiler/blob/master/LICENSE).