DeepSpeech: Can't seem to be able to import gpu version
I’ve created a new Conda environment with python 2.7
Installed all the requirements.
Installed the GPU version of deepspeech via
pip install deepspeech-gpu
Now when I go to bash and run deepspeech -h
I get this:
(deepspeech) nasdin@nasdesktop:~/Documents/github/asaai/speech2text$ deepspeech Traceback (most recent call last): File “/home/nasdin/anaconda3/envs/deepspeech/bin/deepspeech”, line 7, in <module> from deepspeech.client import main File “/home/nasdin/anaconda3/envs/deepspeech/lib/python2.7/site-packages/deepspeech/client.py”, line 10, in <module> from deepspeech.model import Model File “/home/nasdin/anaconda3/envs/deepspeech/lib/python2.7/site-packages/deepspeech/model.py”, line 21, in <module> _model = swig_import_helper() File “/home/nasdin/anaconda3/envs/deepspeech/lib/python2.7/site-packages/deepspeech/model.py”, line 20, in swig_import_helper return importlib.import_module(‘_model’) File “/home/nasdin/anaconda3/envs/deepspeech/lib/python2.7/importlib/init.py”, line 37, in import_module import(name) ImportError: No module named _model
Similarly when I try to import
from deepspeech.model import Model
Got the same error.
When I install the non-gpu version, it works. But I need the gpu version working.
About this issue
- Original URL
- State: closed
- Created 6 years ago
- Comments: 17 (1 by maintainers)
For your SIGSEGV, make sure you are passing all the proper arguments, it’s something several people already reported. Thanks for the CPU infos, it seems that even though your Ryzen has AVX2, something was wrong. Yet we target genric-x86-64, so that’s strange. But I don’t have hardware (nor time) to investigate that further :[
Please try in a fresh and new virtualenv: