Long time waiting in "build ADCME" when "Looking for TensorFlow Dynamic Libraries"

[zhehaoli1999]

Hi there,
when I build ADCME, the building process seems to be stuck by "Looking for TensorFlow Dynamic Libraries"(have waited for 30+ minutes or so),
and the output in REPL and my settings are as follows.

Settings:

• MacBook-Pro.local 20.3.0 Darwin Kernel Version 20.3.0 X86_64
• OS version: MacOS Big Sur 11.2.1
• julia version 1.5.3
• conda 4.8.4 (in dir ~/.julia/adcme/condabin)
• without Tensorflow pre-installed in conda

(@v1.5) pkg> build --verbose ADCME
   Building Conda ─→ `~/.julia/packages/Conda/tJJuN/deps/build.log`
   Building PyCall → `~/.julia/packages/PyCall/tqyST/deps/build.log`
[ Info: PyCall is using /Users/lizhehao/.julia/adcme/bin/python (Python 3.7.7) at /Users/lizhehao/.julia/adcme/bin/python, libpython = /Users/lizhehao/.julia/adcme/lib/libpython3.7m.dylib
[ Info: /Users/lizhehao/.julia/packages/PyCall/tqyST/deps/deps.jl has not changed
[ Info: /Users/lizhehao/.julia/prefs/PyCall has not changed
   Building CMake ─→ `~/.julia/packages/CMake/ULbyn/deps/build.log`
   Building HDF5 ──→ `~/.julia/packages/HDF5/T1b9x/deps/build.log`
   Building FFTW ──→ `~/.julia/packages/FFTW/G3lSO/deps/build.log`
   Building ADCME ─→ `~/.julia/packages/ADCME/hCmEo/deps/build.log`
┌ Warning: Pkg.installed() is deprecated
└ @ Pkg /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.5/Pkg/src/Pkg.jl:554
┌ Warning: Pkg.installed() is deprecated
└ @ Pkg /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.5/Pkg/src/Pkg.jl:554
[ Info: Your Julia version is 1.5.3, current ADCME version is 0.6.7, ADCME dependencies installation path: /Users/lizhehao/.julia/adcme
[ Info:  --------------- (1/6) Install Tensorflow Dependencies  --------------- 
[ Info: ADCME dependencies have already been installed.
[ Info: Installing conda dependencies...
Collecting package metadata (repodata.json): done
Solving environment: done


==> WARNING: A newer version of conda exists. <==
  current version: 4.8.4
  latest version: 4.9.2

Please update conda by running

    $ conda update -n base conda


#
# To activate this environment, use
#
#     $ conda activate base
#
# To deactivate an active environment, use
#
#     $ conda deactivate

[ Info:  --------------- (2/6) Check Python Version  --------------- 
┌ Warning: Pkg.installed() is deprecated
└ @ Pkg /Users/julia/buildbot/worker/package_macos64/build/usr/share/julia/stdlib/v1.5/Pkg/src/Pkg.jl:554
   Building Conda ─→ `~/.julia/packages/Conda/tJJuN/deps/build.log`
   Building PyCall → `~/.julia/packages/PyCall/tqyST/deps/build.log`
┌ Info: PyCall Python version: /Users/lizhehao/.julia/adcme/bin/python
└ Conda Python version: /Users/lizhehao/.julia/adcme/bin/python
[ Info:  --------------- (3/6) Looking for TensorFlow Dynamic Libraries --------------- 

[zhehaoli1999]

There is a related issue: kailaix/ADCME.jl#64
After changing to Julia1.3.1, build ADCME has succeeded.

[zhehaoli1999]

When build AdFem, there is one bug worth mentioning. @kailaix

../CholeskyOp/CholeskyOp.h:1:10: fatal error: 'adept.h' file not found
#include "adept.h"
         ^~~~~~~~~
1 error generated.

Solution: I directly copy all headers from .julia/adcme/lib/Libraries/Adept-2/include into .julia/adcme/include/ and solve this problem.

Now using AdFem works. Hurrah!

[zhehaoli1999]

It's time to close this issue.

[kailaix]

Thanks for pointing it out.