I have been having small issues with Jupyter (I have used and upgraded since very early days of IPython NB) I figured it was time to start fresh. and extensions so I went in and waxed all Jupyter and Ipython files I could find in bin and dist... I reinstalled Jupyter and the notebook came back fine and was running with no errors,, then I installed NB extensions with
sudo pip3 install https://github.com/ipython-contrib/IPython-notebook-extensions/archive/master.zip --user --upgrade
And now on startup I get and I can't load the config page:
tom@tom-sam:~$ jupyter notebook
/usr/local/lib/python3.4/dist-packages/widgetsnbextension/__init__.py:30: UserWarning: To use the jupyter-js-widgets nbextension, you'll need to update
the Jupyter notebook to version 4.2 or later.
the Jupyter notebook to version 4.2 or later.""")
[W 08:24:02.477 NotebookApp] Error loading server extension nbextensions
Traceback (most recent call last):
File "/usr/local/lib/python3.4/dist-packages/notebook/notebookapp.py", line 1042, in init_server_extensions
mod = importlib.import_module(modulename)
File "/usr/lib/python3.4/importlib/__init__.py", line 109, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "<frozen importlib._bootstrap>", line 2254, in _gcd_import
File "<frozen importlib._bootstrap>", line 2237, in _find_and_load
File "<frozen importlib._bootstrap>", line 2226, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 1200, in _load_unlocked
File "<frozen importlib._bootstrap>", line 1129, in _exec
File "<frozen importlib._bootstrap>", line 1471, in exec_module
File "<frozen importlib._bootstrap>", line 321, in _call_with_frames_removed
File "/home/tom/.local/share/jupyter/extensions/nbextensions.py", line 21, in <module>
import yaml
ImportError: No module named 'yaml'
To boot I confirm that notebook is at 4,2
jupyter-notebook --version
4.2.0
FWIW the bulk of issues seem to have started with the install of the IPywidgets 5 which also were/are auto installed by the NB extensions now, wondering if an install I did previously on pip (python2) vs a subsequent pip3 install since Jupyter is running via a pip3 install.. could be some part of the problem? but don't see any evidence of it?
Hi there. Sorry to hear you're having problems. I'm afraid that installation is the biggest problem we have with this repo at the moment, lots of problems many and various. On the plus side, it looks like most the installation has gone ok, and you're just missing the yaml dependency for some reason (our dependency lists are somewhat patchy, and in several places, I think this particular issue results from yaml missing from setup.py dependencies).
Hopefully, this should be fixed by issuing
pip install pyyaml
(or, in your case, presumably sudo pip3 install pyyaml)
(The yaml python package is provided by the PyYAML pip package)
on closer inspection, it seems that setup.py _does_ actually specify pyyaml as a requirement, so I'm not really sure why it hasn't been installed in your case...
Sorry but it is installed so I have no idea why it is not being found..!
Hmm. Curiouser and curiouser. OK, so to clarify, could you let me know:
python3 jupyter notebook or similar?import yaml from a python3/ipython3 prompt?also:
which jupyter giveswhich jupyter-notebook gives1) I normally just do >>jupyter notebook
2)From in a p3 notebook I get no module named yaml when doing 'import yaml'
3) same from a p3 shell
4)which jupyter goves /usr/local/bin/jupyter
5)which jupyter-notebook gives /usr/local/bin/jupyter-notebook
Very weird that yaml won't import seems something messed up with base p3? any ideas?
Very odd. And
pip show pyyaml
shows it to be installed?
If you're on a system using apt, you might try the suggestion in this comment: https://github.com/ipython-contrib/IPython-notebook-extensions/issues/304#issuecomment-169483106
And, to clarify what python is getting used for jupyter, what do you get for head -1which jupyterand `head -1 `which jupyter-notebook? (I'm thinking of shebang lines, but I'd expect them to point to the python used for the most recently installed jupyter)
Are you using any virtual environments?
very strange pip3 show pyyaml just returns to command prompt then using pip3 list I don't see any pyyaml (or yaml)
Not using vitrual env :-(
tom@tom-sam:~$ head -1 which jupyter
tom@tom-sam:~$ head -1 which jupyter-notebook
OK, and what does the pip3 install pyyaml give? If pip3 is claiming simultaneously to have installed pyyaml but also not showing it as installed, then I think this is more of a pip problem!
hunh pip3 just uninstalled PyYAML and reinstalled and that fixed that one!
hahaha ok! So probably an issue somewhere with pyyaml and/or pip...
and now have the nbextensions page! Thanks for all the help!
no worries, you're welcome 馃槃
still getting this erroneous warning but don't think it is bad? /usr/local/lib/python3.4/dist-packages/widgetsnbextension/init.py:30: UserWarning: To use the jupyter-js-widgets nbextension, you'll need to update
the Jupyter notebook to version 4.2 or later.
the Jupyter notebook to version 4.2 or later.""")
yeah, that seems strange, given you seem to be running 4.2 already anyway... It doesn't sound like a problem, and as far as I'm aware, it's not used by anything in this repo anyway. Based on the experience with pyyaml, you might try a
pip uninstall/resintall of ipywidgets... 馃槅
You might want to check your kernelspec. Juyper loads the Python version defined there:
hasch@VirtualMint ~ $ jupyter kernelspec list
Available kernels:
python3 /home/hasch/.local/share/jupyter/kernels/python3
hasch@VirtualMint ~ $ cat ~/.local/share/jupyter/kernels/python3/kernel.json
{
"language": "python",
"argv": [
"/home/hasch/anaconda3/bin/python",
"-m",
"ipykernel",
"-f",
"{connection_file}"
],
"display_name": "Python 3"
This often leads to confusion. I usually remove the absolute path and only use python
The warning about updating to jupyter 4.2 seems to fire regardless of the actually installed version, as noted in #591
closing this, as it seems it was resolved ok.
For followers, I was able to work around this by doing a "conda activate xx" first, then it worked.
Requirement already satisfied: pyyaml in /home/ad/venv/lib/python3.5/site-packages (3.13)
File "/tmp/easy_install-Q6yl7W/models-0.9.3/setup.py", line 25, in
File "/tmp/easy_install-Q6yl7W/models-0.9.3/models/__init__.py", line 24, in
File "/tmp/easy_install-Q6yl7W/models-0.9.3/models/props.py", line 23, in
ImportError: No module named yaml
the issue got resolved for centos using below command.
pip install PyYAML
I solve it by
pip install pyyaml==5.1.0
the newest is 5.3.0, it doesn't work
pip install pyyaml --ignore-installed did the trick.
pip install pyyaml --ignore-installeddid the trick.
Thanks. I've tried the install pyaml method and got the same error. This fix did the trick for me.
for me, none of it works;
``` $ pip freeze
python-dateutil==2.8.1
PyYAML==5.3.1
ruamel.yaml==0.16.12
ruamel.yaml.clib==0.2.2
six==1.15.0
strictyaml==1.1.1
but when I run my script:
import yaml
ModuleNotFoundError: No module named 'yaml'
```
I have the exact same issue as jurgenweber. I also tried pip uninstall pyymal to uninstall v5.3.1 and running pip install PyYAML==5.1.0 as per starxchina's suggestion, but that didn't seem to make a difference. I also tried v3.13, but even that old of a version didn't seem to work for me. I'm trying this on Windows.
@MyNameIsTrez I eventually got it to work, but I am unsure what did it. I think just opening up in a new shell was part of it, it was all very weird and frustrating on Friday. :0
@jurgenweber That didn't fix the issue for me in VS Code's shell, but let's hope it works for someone else.
Most helpful comment
Hi there. Sorry to hear you're having problems. I'm afraid that installation is the biggest problem we have with this repo at the moment, lots of problems many and various. On the plus side, it looks like most the installation has gone ok, and you're just missing the yaml dependency for some reason (our dependency lists are somewhat patchy, and in several places, I think this particular issue results from yaml missing from setup.py dependencies).
Hopefully, this should be fixed by issuing
(or, in your case, presumably
sudo pip3 install pyyaml)