TensorflowEmitter has not supported operator [batch_normalization].

[dhaase-de]

This happened to me when converting a Caffe classification model using batch norm to TensorFlow. I used Python 3 and MMdnn-0.1.1, installed via pip.

Should be fixed by changing mmdnn/conversion/caffe/mapper.py:193 from
return Node.create('batch_normalization', **kwargs) to
return Node.create('BatchNorm', **kwargs).

[kitstar]

Fixed in master branch. But the batch normalization of caffe converter is not well tested. Glad if you can help with it. Thanks.