Closed Hierosme closed 2 years ago
Hello,
I'd like to report the same issue with Ubuntu 21.10 . It looks like the main problem is that there are no pre-compiled llvmlite
binaries for Python 3.9 (current default) and llvmlite==0.31
needs LLVM==8 for compilation, which is not available (9 and 13 are).
I Installed Python 3.8 (by this recipe) and everything looks OK again. I still need to test a few more things but it looks promising.
Update: Still there is some TensorFlow bug with "Another metric with the same name..." According to https://discuss.tensorflow.org/t/alreadyexistserror-another-metric-with-the-same-name-already-exists/3787, Python 3.7.9 might be a better choice.
See below the installation recipe I used last time.
From my readme:
Install the package with the following command. This way, you'll have
all useful "binaries" (e.g. tensorflow-tts-preprocess
) and any changes
to the code will take effect immediately as well.
pip install --upgrade --editable .
Continue with installing german_transliterate
(from dockerfile
):
pip install git+https://github.com/repodiac/german_transliterate.git#egg=german_transliterate
Then, I only needed to fix numpy
and h5py
versions:
pip install --upgrade numpy~=1.19.2
pip install --upgrade h5py~=3.1.0
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.
Hello,
pip install Instructions not work at all due to llvm installation and unfortunally after one hour of researh i simply not understand how install TensorFlowTTS.
LLVM version... 11.0.1 .... RuntimeError: Building llvmlite requires LLVM 7.0+ Be sure to set LLVM_CONFIG to the right executable path. Read the documentation at http://llvmlite.pydata.org/ for more information about building llvmlite.
I have follow llvm installation instruction without succes
Anyone have test installation on Debian 11 ? Someone know if documentation better as a simple 'pip install' exist ?
TensorFlowTTS installation look ultra complexe and the documentation don't reflet that ....
Regards