Deeplabcut with GPU support

I’m trying to run deeplabcut with GPU support using Anaconda.

I was able to install tensorflow, but every time that I try to import deeplabcut it reports an error, I’m going to copy here what it says.

Can you help me? I’m not familiar with coding.

Thank you in advance

In [1]: import deeplabcut


ImportError                               Traceback (most recent call last)

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\python\ in <module>()


---> 58   from tensorflow.python.pywrap_tensorflow_internal import *

     59   from tensorflow.python.pywrap_tensorflow_internal import __version__

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\python\ in <module>()

     27             return _mod

---> 28     _pywrap_tensorflow_internal = swig_import_helper()

     29     del swig_import_helper

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\python\ in swig_import_helper()

     23             try:

---> 24                 _mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname, description)

     25             finally:

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\ in load_module(name, file, filename, details)

    242         else:

--> 243             return load_dynamic(name, filename, file)

    244     elif type_ == PKG_DIRECTORY:

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\ in load_dynamic(name, path, file)

    342             name=name, loader=loader, origin=path)

--> 343         return _load(spec)


ImportError: DLL load failed: The specified procedure could not be found.

During handling of the above exception, another exception occurred:

ImportError                               Traceback (most recent call last)

<ipython-input-1-52d2df9c7fa5> in <module>()

----> 1 import deeplabcut

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\deeplabcut\ in <module>()

     27     else:

     28         mpl.use('Agg')

---> 29     from deeplabcut import generate_training_dataset

     30     from deeplabcut import refine_training_dataset

     31     from deeplabcut.generate_training_dataset import label_frames, dropannotationfileentriesduetodeletedimages, comparevideolistsanddatafolders

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\deeplabcut\generate_training_dataset\ in <module>()

     17 else:

     18     from deeplabcut.generate_training_dataset.auxfun_drag_label import *

---> 19     from deeplabcut.generate_training_dataset.labeling_toolbox import *

     20     from deeplabcut.generate_training_dataset.multiple_individual_labeling_toolbox import *

     21     from deeplabcut.generate_training_dataset.frame_extraction_toolbox import *

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\deeplabcut\generate_training_dataset\ in <module>()

     25 import argparse

     26 from deeplabcut.generate_training_dataset import auxfun_drag_label

---> 27 from deeplabcut.utils import auxiliaryfunctions

     28 from matplotlib.backends.backend_wxagg import FigureCanvasWxAgg as FigureCanvas

     29 from matplotlib.backends.backend_wxagg import NavigationToolbar2WxAgg as NavigationToolbar

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\deeplabcut\utils\ in <module>()

----> 1 from deeplabcut.utils.make_labeled_video import *

      2 from deeplabcut.utils.auxiliaryfunctions import *

      3 from deeplabcut.utils.video_processor import *

      4 from deeplabcut.utils.plotting import *


c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\deeplabcut\utils\ in <module>()


     37 from deeplabcut.utils import auxiliaryfunctions

---> 38 from deeplabcut.pose_estimation_tensorflow.config import load_config

     39 from skimage.util import img_as_ubyte

     40 from skimage.draw import circle_perimeter, circle, line,line_aa

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\deeplabcut\pose_estimation_tensorflow\ in <module>()

     11 from deeplabcut.pose_estimation_tensorflow.dataset import *

     12 from deeplabcut.pose_estimation_tensorflow.models import *

---> 13 from deeplabcut.pose_estimation_tensorflow.nnet import *

     14 from deeplabcut.pose_estimation_tensorflow.util import *


c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\deeplabcut\pose_estimation_tensorflow\nnet\ in <module>()


     13 """

---> 14 from deeplabcut.pose_estimation_tensorflow.nnet.losses import *

     15 from deeplabcut.pose_estimation_tensorflow.nnet.net_factory import *

     16 from deeplabcut.pose_estimation_tensorflow.nnet.pose_net import *

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\deeplabcut\pose_estimation_tensorflow\nnet\ in <module>()


      4 '''

----> 5 import tensorflow as tf


      7 from tensorflow.python.ops import math_ops

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\ in <module>()


     23 # pylint: disable=g-bad-import-order

---> 24 from tensorflow.python import pywrap_tensorflow  # pylint: disable=unused-import


     26 from tensorflow._api.v1 import app

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\python\ in <module>()

     47 import numpy as np


---> 49 from tensorflow.python import pywrap_tensorflow


     51 # Protocol buffers

c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\python\ in <module>()

     72 for some common reasons and solutions.  Include the entire stack trace

     73 above this error message when asking for help.""" % traceback.format_exc()

---> 74   raise ImportError(msg)


     76 # pylint: enable=wildcard-import,g-import-not-at-top,unused-import,line-too-long

ImportError: Traceback (most recent call last):

  File "c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\python\", line 58, in <module>

    from tensorflow.python.pywrap_tensorflow_internal import *

  File "c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\python\", line 28, in <module>

    _pywrap_tensorflow_internal = swig_import_helper()

  File "c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\site-packages\tensorflow\python\", line 24, in swig_import_helper

    _mod = imp.load_module('_pywrap_tensorflow_internal', fp, pathname, description)

  File "c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\", line 243, in load_module

    return load_dynamic(name, filename, file)

  File "c:\users\vpaschoal\.conda\envs\dlc-windowsgpu\lib\", line 343, in load_dynamic

    return _load(spec)

ImportError: DLL load failed: The specified procedure could not be found.

Failed to load the native TensorFlow runtime.


for some common reasons and solutions.  Include the entire stack trace

above this error message when asking for help.​

Welcome to the forum, @vpaschoal! :smile:

From the error message, it looks like a TensorFlow installation issue. Issues such as this one suggest that you need to install tensorflow-gpu into your conda environment, and ensure your CUDA version matches what the package needs. Did you carefully follow the TensorFlow installation instructions for conda?

I did, I was able to install tensorflow-gpu successfully onto the conda environment, and the CUDA should match the package needs (maybe a bit above really). I thought it was something with the commands while trying to load deeplabcut that was not working.

And I saw that post, but I am already using Anaconda, and the CUDA version is the same between the computer and anaconda…

1 Like

Thanks for the reply, @vpaschoal. In that case, maybe @MWMathis has some ideas how to troubleshoot further?

It does seem to be a tensorflow issue; what happens when you import tensorflow

1 Like