Closed mrgloom closed 5 years ago
When I run ./run_demo.sh under merlin/egs/slt_arctic/s1 I get:
./run_demo.sh
merlin/egs/slt_arctic/s1
./run_demo.sh Step 1: data is ready! Merlin default voice settings configured in conf/global_settings.cfg setup done...! Step 2: preparing config files for acoustic, duration models... Duration configuration settings stored in conf/duration_slt_arctic_demo.conf Acoustic configuration settings stored in conf/acoustic_slt_arctic_demo.conf preparing config files for synthesis... Duration configuration settings stored in conf/test_dur_synth_slt_arctic_demo.conf Acoustic configuration settings stored in conf/test_synth_slt_arctic_demo.conf Step 3: training duration model... Architecture: x86_64 /Users/user/external_projects/text-to-speech/merlin/src/setup_env.sh: line 21: lsb_release: command not found Distribution: HOSTNAME=user-MacBook-Pro.local USER=user PATH: /usr/local/bin /usr/bin /bin /usr/sbin /sbin /opt/X11/bin LD_LIBRARY_PATH: PYTHONPATH: /Users/user/external_projects/text-to-speech PYTHONBIN: python MERLIN_THEANO_FLAGS: cuda.root=/usr/local/8.0 floatX=float32 on_unused_input=ignore No GPU is available! Running on CPU... Traceback (most recent call last): File "/Users/user/external_projects/text-to-speech/merlin/src/run_merlin.py", line 56, in <module> import theano ImportError: No module named theano
But I can import theano from python:
import theano theano.__file__ '/usr/local/lib/python3.6/site-packages/theano/__init__.py'
Installing under virtualenv helped:
python3 -m virtualenv merlin-env source merlin-env/bin/activate pip install -r requirements.txt
When I run
./run_demo.sh
undermerlin/egs/slt_arctic/s1
I get:But I can import theano from python: