I did conda update --all today
and now hvplot
seems broken. Anybody recognize this problem?
(pangeo) rsignell@Elio:~$ conda list | grep -E "bokeh|pyviz|widgets|hvplot|geoviews|holoviews|panel"
bokeh 3.4.1 pyhd8ed1ab_0 conda-forge
geoviews 1.12.0 pyhd8ed1ab_0 conda-forge
geoviews-core 1.12.0 pyha770c72_0 conda-forge
holoviews 1.18.3 pyhd8ed1ab_0 conda-forge
hvplot 0.10.0 pyhd8ed1ab_0 conda-forge
ipywidgets 8.1.3 pyhd8ed1ab_0 conda-forge
jupyter-panel-proxy 0.1.0 py_0 conda-forge
jupyter_bokeh 4.0.4 pyhd8ed1ab_0 conda-forge
jupyterlab_widgets 3.0.11 pyhd8ed1ab_0 conda-forge
panel 1.4.3 pyhd8ed1ab_0 conda-forge
pyviz_comms 3.0.2 pyhd8ed1ab_1 conda-forge
widgetsnbextension 4.0.11 pyhd8ed1ab_0 conda-forge
Just updated everything to the same versions as you and updated conda. I have no issues, but I also don’t have jupyter-panel-proxy
installed. My understanding is the proxy server is only launched when you try to visit the corresponding endpoint, but is it possible there are some other interactions?
I tried removing jupyter-panel-proxy
but still have same problem.
It’s not clear to me from that issue what packages I should try downgrading. Do you know?
These are my versions:
bokeh 3.4.0 pypi_0 pypi
geoviews 1.12.0 pypi_0 pypi
holoviews 1.18.3 pypi_0 pypi
hvplot 0.10.0 pypi_0 pypi
ipywidgets 8.1.1 pyhd8ed1ab_0 conda-forge
ipywidgets_bokeh 1.5.0 py_0 bokeh
jupyter-bokeh 4.0.4 pypi_0 pypi
jupyterlab_widgets 3.0.9 pyhd8ed1ab_0 conda-forge
panel 1.4.0a3.post201+gdefe76aa.dirty pypi_0 pypi
panel-chat-examples 0.0.0 pypi_0 pypi
pyviz_comms 2.3.2 py_0 pyviz/label/dev
widgetsnbextension 4.0.9 pyhd8ed1ab_0 conda-forge
Try downgrading Panel to 1.4.2.
I tried downgrading panel to 1.4.2, but didn’t work.
And I’m confused: I’m only using hvplot – why is panel even involved?
Aha! Downgrading to pyviz_comms==2.3.2
fixed the problem for me. Thanks @ahuang11 !
Hoxbro
May 28, 2024, 3:46pm
10
Panel handles the communication between Python and Jupyter. Also what makes the widgets.
Ah, okay! While we are at it, what does pyviz_comms do? That sounds like communication also!
Hmm, testing Jupyter lab and downgrading pyviz_comms doesn’t seem to work for me.
After testing a lot of package variations, pinning pip install panel!=1.4.3
works.