-
Notifications
You must be signed in to change notification settings - Fork 3.4k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[FRONTEND][MXNET] L2Normalization is not supported in nnvm #1223
Comments
@PariksheetPinjari909 i see. But it seem to be an error during test phase. I hope to see this feature as soon as possible |
Merged
I just built TVM from source, yet I am getting the same error (L2Normalization is not supported in nnvm). My model is from mxnet. What symbol version is needed for this to work? |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi all. I got this error when i run model VGG_VOC0712_SSD_300x300 in mxnet through command
nnvm_sym, nnvm_params = nnvm.frontend.from_mxnet(mx_sym, args, auxs)
Error is:
File "/home/prdcv181/PycharmProjects/darknet_nnvm/Phase2/from_mxnet.py", line 86, in
nnvm_sym, nnvm_params = nnvm.frontend.from_mxnet(mx_sym, args, auxs)
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 371, in from_mxnet
sym = _from_mxnet_impl(symbol, {})
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 334, in _from_mxnet_impl
childs = [_from_mxnet_impl(childs[i], graph) for i in range(len(childs.list_outputs()))]
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 336, in _from_mxnet_impl
node = _convert_symbol(op_name, childs, attr)
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 295, in _convert_symbol
_raise_not_supported('Operator: ' + op_name)
File "/usr/local/lib/python2.7/dist-packages/nnvm-0.8.0-py2.7.egg/nnvm/frontend/mxnet.py", line 24, in _raise_not_supported
raise NotImplementedError(err)
NotImplementedError: Operator: L2Normalization is not supported in nnvm.
I checked that mxnet in nnvm now only support those layer:
'Activation' : _activations,
'BatchNorm' : _batch_norm,
'BatchNorm_v1' : _batch_norm,
'Cast' : _rename('cast'),
'Concat' : _concat,
'Convolution' : _conv2d,
'Convolution_v1': _conv2d,
'Deconvolution' : _conv2d_transpose,
'Dropout' : _dropout,
'Flatten' : _rename('flatten'),
'FullyConnected': _dense,
'LeakyReLU' : _leaky_relu,
'Pooling' : _pooling,
'Pooling_v1' : _pooling,
'Reshape' : _reshape,
'SliceChannel' : _split,
'split' : _split,
'Softmax' : _rename('softmax'),
'SoftmaxOutput' : _softmax_output,
'concat' : _concat,
'max_axis' : _rename('max'),
'min_axis' : _rename('min'),
'reshape' : _reshape,
'sum_axis' : _rename('sum'),
'UpSampling' : _upsampling,
'clip' : _clip
So, how can i work around with this issue?
The text was updated successfully, but these errors were encountered: