I have created this Conda environment:
conda env create -f environment.yml
The environment.yml file:
name: deep3d_pytorch
channels:
  - pytorch
  - conda-forge
  - defaults
dependencies:
  - python=3.6
  - pytorch=1.6.0
  - torchvision=0.7.0
  - numpy=1.18.1
  - scikit-image=0.16.2
  - scipy=1.4.1
  - pillow=6.2.1
  - pip
  - ipython=7.13.0
  - yaml=0.1.7
  - pip:
    - matplotlib==2.2.5
    - opencv-python==3.4.9.33
    - tensorboard==1.15.0
    - tensorflow==1.15.0
    - kornia==0.5.5
    - dominate==2.6.0
    - trimesh==3.9.20
I activate the Conda environment. But even a simple statement like python -c "import torch; print(torch.__version__)" to get the PyTorch version throws the undefined symbol error:
(deep3d_pytorch) m3@i7:~/repos/Deep3DFaceRecon_pytorch> python -c "import torch; print(torch.__version__)"
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/home/m3/anaconda3/envs/deep3d_pytorch/lib/python3.6/site-packages/torch/__init__.py", line 189, in <module>
    from torch._C import *
ImportError: /home/m3/anaconda3/envs/deep3d_pytorch/lib/python3.6/site-packages/torch/lib/libtorch_cpu.so: undefined symbol: iJIT_IsProfilingActive
I believe the PyTorch installed by Conda is broken. But the Conda logs are all fine. Does anyone have a clue or hint? I'm receiving the undefined symbol error on both my local machine and on Google Colab.
Even a minimal Environment like below, would throw similar errors:
conda create -n minimal_pytorch python=3.6 pytorch torchvision torchaudio -c pytorch
source activate minimal_pytorch && python -c "import torch; print(torch.__version__)"
A similar undefined symbol error is thrown:
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/usr/local/envs/minimal_pytorch/lib/python3.6/site-packages/torch/__init__.py", line 197, in <module>
    from torch._C import *  # noqa: F403
ImportError: /usr/local/envs/minimal_pytorch/lib/python3.6/site-packages/torch/lib/libtorch_cpu.so: undefined symbol: iJIT_NotifyEvent
When Python version is omitted while creating the environment:
conda create -n minimal_pytorch python pytorch torchvision torchaudio -c pytorch
The error is resolved:
source activate minimal_pytorch && python -c "import torch; print(torch.__version__)"
PyTorch version is received without any error:
2.2.2
                Downgrade mkl package to an earlier version. It should work.
Using pip install instead of conda install solved the problem for me:
pip install torch==x.x+cu11x torchvision==x.x.0+cu11x torchaudio==x.x --extra-index-url https://download.pytorch.org/whl/cu11x
                        If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With