MMdnn is a set of tools to help users inter-operate among different deep learning frameworks. E.g. model conversion and visualization. Convert models between Caffe, Keras, MXNet, Tensorflow, CNTK, PyTorch Onnx and CoreML.
Python version: Python 3.5.4 (v3.5.4:3f56838, Aug 8 2017, 02:17:05) [MSC v.1900 64 bit (AMD64)] on win32
1) I installed MMdnn with pip install mmdnn as instructed, however notice this does NOT install the required mxnet package that must be installed separately (I do not know if this is intended).
2) I am trying to convert models from MXNET into Keras. I tried with the VGG16 example as shown below, but I get an assertion failure and conversion stops:
D:\Programs\WinPython\MMdnn\scripts>mmdownload -f mxnet -n vgg16 -o D:\Users\mzatt\Projects\ModelZoo
...
MXNet Model vgg16 saved as [D:\Users\mzatt\Projects\ModelZoo\vgg16-symbol.json] and [D:\Users\mzatt\Projects\ModelZoo\vgg16-0000.params].
D:\Programs\WinPython\MMdnn\scripts>mmconvert -sf mxnet -iw D:\Users\mzatt\Projects\ModelZoo\vgg16-0000.params -in D:\Users\mzatt\Projects\ModelZoo\vgg16-symbol.json -df keras -om D:\Users\mzatt\Projects\ModelZoo\vgg16.h5
[09:51:28] C:\Jenkins\workspace\mxnet-tag\mxnet\src\nnvm\legacy_json_util.cc:209: Loading symbol saved by previous version v0.8.0. Attempting to upgrade...
[09:51:28] C:\Jenkins\workspace\mxnet-tag\mxnet\src\nnvm\legacy_json_util.cc:217: Symbol successfully upgraded!
D:\Programs\WinPython\MMdnn\python-3.5.4.amd64\lib\site-packages\mxnet\module\base_module.py:55: UserWarning: [91mYou created Module with Module(..., label_names=['softmax_label']) but input with name 'softmax_label' is not found in symbol.list_arguments(). Did you mean one of:
data
prob_label[0m
warnings.warn(msg)
Traceback (most recent call last):
File "D:\Programs\WinPython\MMdnn\python-3.5.4.amd64\lib\runpy.py", line 193, in _run_module_as_main
"__main__", mod_spec)
File "D:\Programs\WinPython\MMdnn\python-3.5.4.amd64\lib\runpy.py", line 85, in _run_code
exec(code, run_globals)
File "D:\Programs\WinPython\MMdnn\python-3.5.4.amd64\Scripts\mmconvert.exe\__main__.py", line 9, in <module>
File "D:\Programs\WinPython\MMdnn\python-3.5.4.amd64\lib\site-packages\mmdnn\conversion\_script\convert.py", line 102, in _main
ret = convertToIR._convert(ir_args)
File "D:\Programs\WinPython\MMdnn\python-3.5.4.amd64\lib\site-packages\mmdnn\conversion\_script\convertToIR.py", line 80, in _convert
parser = MXNetParser(model)
File "D:\Programs\WinPython\MMdnn\python-3.5.4.amd64\lib\site-packages\mmdnn\conversion\mxnet\mxnet_parser.py", line 242, in __init__
assert isinstance(input_arg[3], list)
AssertionError
Platform: WinPython under Windows 10
Python version: Python 3.5.4 (v3.5.4:3f56838, Aug 8 2017, 02:17:05) [MSC v.1900 64 bit (AMD64)] on win32
1) I installed MMdnn with pip install mmdnn as instructed, however notice this does NOT install the required mxnet package that must be installed separately (I do not know if this is intended).
2) I am trying to convert models from MXNET into Keras. I tried with the VGG16 example as shown below, but I get an assertion failure and conversion stops:
Below the list of my installed packages: