An error occurred while compiling the source code

k2-fsa / fast_rnnt

A torch implementation of a recursion which turns out to be useful for RNN-T.

Other

139 stars 22 forks source link

Closed jiay7 closed 2 years ago

jiay7 commented 2 years ago

Thank you for the fast_rnnt, I got the error (show blow) when I run "python setup.py install". 1655099928(1)

python=3.8.11 torch verison=1.10.1 cudatoolkit=10.2.89 CUDA version=10.2. GCC version =5.3.1. cmake version=3.23.0

Is this known issue? How can it be debugged and solved?

Thank you!

csukuangfj commented 2 years ago

It requires a version of GCC that supports C++14.

From https://gcc.gnu.org/onlinedocs/libstdc++/manual/status.html#status.iso.2014, GCC >= 6.1 supports C++14. Your GCC version 5.3 is too low. Please try a newer one, e.g., GCC 7.5

Note: You can also try to install k2: https://k2-fsa.github.io/k2/installation/index.html , which also contains pruned RNN-T. You can find its usage in https://github.com/k2-fsa/icefall

jiay7 commented 2 years ago

Thank you for your reply. I used gcc==7.5, which works.