Skip to content
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

fold incompatible with TF-gpu 1.10.1 #99

Open
bjtho08 opened this issue Sep 24, 2018 · 1 comment
Open

fold incompatible with TF-gpu 1.10.1 #99

bjtho08 opened this issue Sep 24, 2018 · 1 comment

Comments

@bjtho08
Copy link

bjtho08 commented Sep 24, 2018

Installing tensorflow-fold in python alongside tensorflow-gpu 1.10.1 results in the following error traceback:

---------------------------------------------------------------------------
NotFoundError                             Traceback (most recent call last)
<ipython-input-2-5f32c6c0a67b> in <module>()
      3 import tensorflow as tf
      4 sess = tf.InteractiveSession()
----> 5 import tensorflow_fold as td

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/__init__.py in <module>()
     20 
     21 # pylint: disable=wildcard-import, unused-import
---> 22 from tensorflow_fold.blocks.block_compiler import *
     23 from tensorflow_fold.blocks.blocks import *
     24 from tensorflow_fold.blocks.layers import *

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/blocks/block_compiler.py in <module>()
     26 from six.moves import xrange  # pylint: disable=redefined-builtin
     27 import tensorflow as tf
---> 28 from tensorflow_fold.blocks import loom_ops
     29 from tensorflow_fold.blocks import result_types as tdt
     30 from tensorflow_fold.blocks import util

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/blocks/loom_ops.py in <module>()
     17 from __future__ import print_function
     18 # import google3
---> 19 import tensorflow_fold.blocks.result_types as tdt
     20 from tensorflow_fold.public import loom
     21 

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/blocks/result_types.py in <module>()
     25 from six.moves import zip  # pylint: disable=redefined-builtin
     26 import tensorflow as tf
---> 27 from tensorflow_fold.public import loom
     28 
     29 

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/public/loom.py in <module>()
     15 
     16 # pylint: disable=wildcard-import, unused-import
---> 17 from tensorflow_fold.loom.loom import *
     18 # pylint: enable=wildcard-import, unused-import

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/loom/__init__.py in <module>()
     15 
     16 # pylint: disable=wildcard-import, unused-import
---> 17 from tensorflow_fold.loom.loom import *
     18 # pylint: enable=wildcard-import, unused-import

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/loom/loom.py in <module>()
     34 import tensorflow as tf
     35 
---> 36 from tensorflow_fold.loom import deserializing_weaver_op
     37 from tensorflow_fold.loom import loom_pb2
     38 from tensorflow_fold.loom import pywrapweaver

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/loom/deserializing_weaver_op.py in <module>()
     32 
     33 _deserializing_weaver = tf.load_op_library(os.path.join(
---> 34     tf.resource_loader.get_data_files_path(), '_deserializing_weaver_op.so'))
     35 deserializing_weaver = _deserializing_weaver.deserializing_weaver
     36 

~/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow/python/framework/load_library.py in load_op_library(library_filename)
     54     RuntimeError: when unable to load the library or get the python wrappers.
     55   """
---> 56   lib_handle = py_tf.TF_LoadLibrary(library_filename)
     57 
     58   op_list_str = py_tf.TF_GetOpList(lib_handle)

NotFoundError: /home/bjarne/.pyenv/versions/3.6.0/lib/python3.6/site-packages/tensorflow_fold/loom/_deserializing_weaver_op.so: undefined symbol: _ZTIN10tensorflow8OpKernelE

After a bit of reading on similar errors, it seems the error is due to a compatibility issue. I should note, that all packages have been installed using pip, so there's no compilation error on my part.

I'm running Ubuntu 16.04, Python 3.6.0, CUDA 9.0, and cuDNN 7.2.1.

@ssnvxia
Copy link

ssnvxia commented Dec 12, 2018

the same error

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants