A very similar problem, but different solution.
When I try to do import speech_recognition as sr in Jupyter, it returns the no moduled named... error.
But doing it in a terminal is fine. So I looked through the sys.paths of both. In Jupyter:
//anaconda/lib/python3.5/site-packages/six-1.10.0-py3.5.egg',
'/Users/don',
'/Users/don/Users/don/Documents/PyModules/skynet',
'//anaconda/lib/python35.zip',
'//anaconda/lib/python3.5',
'//anaconda/lib/python3.5/plat-darwin',
'//anaconda/lib/python3.5/lib-dynload',
'/anaconda/lib/python3.5/site-packages/Sphinx-1.3.1-py3.5.egg',
'//anaconda/lib/python3.5/site-packages',
'//anaconda/lib/python3.5/site-packages/aeosa',
'//anaconda/lib/python3.5/site-packages/IPython/extensions',
'/Users/don/.ipython'
In terminal:
/Users/don/Applications/miniconda3/lib/python3.5/site-packages
/datashader-0.4.0-py3.5.egg
/Users/don
/Users/don/Users/don/Documents/PyModules/skynet
/Users/don/Applications/miniconda3/lib/python35.zip
/Users/don/Applications/miniconda3/lib/python3.5
/Users/don/Applications/miniconda3/lib/python3.5/plat-darwin
/Users/don/Applications/miniconda3/lib/python3.5/lib-dynload
/Users/don/Applications/miniconda3/lib/python3.5/site-packages
/Users/don/Applications/miniconda3/lib/python3.5/site-packages/setuptools-23.0.0-py3.5.egg
If I run this first in Jupyter, then the module speech_recognition imports fine:
sys.path.append('/Users/don/Applications/miniconda3/lib/python3.5/site-packages')
I'd just like to know what's going on? I've found a solution of sorts, but I'd like a more elegant one, so I don't have to do sys.path.append every time. This is particularly endemic to a bunch of modules I've installed lately using pip install <module>.
