I read on the blog about plotly support, but it seems to be working only with the IPython kernel with no autocompletion feature, also the IPython widgets are not available even with this kernel. Is there a plan to implement some kind of datalore widgets?
Plotly offers a powerful interface through FigureWidget, which as it’s based on ipywidgets it is also not present here. Will this be supported in the future? At least for the time being colab offers this interface with all the interactivity, autocompletion and docs, this would make a nice adition to datalore at least for the people using notebooks mainly for visualization and prototyping.
Thanks for your suggestion, we definitely have plans both to support plotly as well as other popular visualization libs in Datalore kernel and to support IPython widgets. Will keep you updated on the progress
So it looks like Plotly support now works, which is great, but it is a shame you need to actually run them in Datalore to see the visualisations (rather than export the complete notebook from Pycharm and see the vis straight away)
Some of ipywidgets already work (like IntSlider or Dropdown), but not all of them.
I’ve come across the following issue:
from IPython.display import HTML
import ipywidgets as widgets
out = widgets.Output()
out.append_display_data(HTML("<em>Lorem ipsum</em>"))
out
This code works just fine in the Jupyter notebook, but in Datalore with IPython kernel it immediately says:
A Jupyter widget could not be displayed because the widget state could not be found.
This could happen if the kernel the widget was created in is no longer available,
or if the widget state was not saved in the notebook.
You may be able to create the widget by running the appropriate cells.
And it doesn’t provide any other output (the same thing happens with Datalore kernel). It seems that exactly this widget still doesn’t work.