For questions, use https://discourse.holoviz.org/ instead! This chat is for informal discussions and debugging sessions with developers of the HoloViz.org tools (HoloViews, Datashader, Panel, hvPlot, GeoViews, Param, etc.), but user questions should go to the HoloViz Discourse site so that others won't have to ask them again.
Also try clicking some of the apps to get a feel for the look and feel that Streamlit provides.
At least providing something even simpler based on the Heroku description here https://dash.plotly.com/deployment could be an idea and a huge win I believe.
https://panel.holoviz.org/user_guide/Server_Deployment.html I think panel already does show how to deploy on Heroku here
Anyone else have issues with bokeh after creating a fresh env (as of yesterday)? I'm getting
(unknown) [bokeh] JS/Python version mismatch
(unknown) [bokeh] Library versions: JS (2.2.1) / Python (2.2.3)Apologies if this is the wrong place, but the Bokeh gitter doesn't seem to be a thing anymore.
I don't know where/what needs to be updated to update the JS library
@MarcSkovMadsen , hosting public apps in that way has been discussed as a possibility by Anaconda for the entire time I've been working there, but nothing concrete has happened on that front.
I agree (and have argued internally) that it would be a great idea.
The thing I don't understand is why Heroku, AWS, Azure, Google Cloud and other cloud providers don't provide the same simple 3 step technical solution.
Streamlit is anticipating a revenue stream, whereas Anaconda is not expecting any money to flow from Panel.
The real question is whether Streamlit's revenue model will support their deployment server fully, in the long run. I don't know enough about it to have an opinion there.
Anaconda can definitely provide support for Panel, with an hourly rate for work done.
I assume that the VCs who invested in them believe in their revenue model. Panel is not VC funded.
The thing I am in doubt about is the revenue model. Because the Enterprises I know about prefers being in control. And they have the resources to configure CI/ CD and run Streamlit, Voila, Panel, Dash etc. in their own environments. Why buy that service when it's so easy for DevOps to do themselves?
And that they are doing it already in many cases, anyway.
But getting a brand and connections to a huge community of data scientists with 0-2 years of experience can be very valuable I think. Those data scientist will be the decision takers tomorrow.
Why does images now upload as image links instead of an image I can view directly?
Data App Frameworks are getting a place. Thats what I would say. I think Panel deserves to be there. But I doubt it will ever get there. Not because of the technology but because of the support behind it.
I assure you, all these things are discussed extensively within Anaconda, but so far the HoloViz tools have been primarily developed as ways that our customers can achieve great things, as funded directly by those customers, without necessarily requiring alignment between HoloViz's future and Anaconda's commercial activities. If such alignment does occur, I can easily imagine launching a similar service, and as mentioned doing so has been regularly proposed and considered for many years. So far it hasn't lined up with the roadmap of our commercial activities, but that's not to say it won't next month or next year. That said, other tools may already have deep inroads into mindshare at that point, which is a risk we all take.
Yes, putting a statement like that on our sites, on balance, probably will be a good idea. I'll try to draft something.
~/anaconda3/envs/py3/lib/python3.7/site-packages/holoviews/plotting/plot.py in compute_ranges(self, obj, key, ranges)
633 if (not (axiswise and not isinstance(obj, HoloMap)) or
634 (not framewise and isinstance(obj, HoloMap))):
--> 635 self._compute_group_range(group, elements, ranges, framewise, self.top_level)
636 self.ranges.update(ranges)
637 return ranges
~/anaconda3/envs/py3/lib/python3.7/site-packages/holoviews/plotting/plot.py in _compute_group_range(cls, group, elements, ranges, framewise, top_level)
786 values = el.nodes.dimension_values(2, expanded=False)
787 else:
--> 788 values = el.dimension_values(el_dim, expanded=False)
789 elif isinstance(el, Graph) and el_dim in el.nodes:
790 values = el.nodes.dimension_values(el_dim, expanded=False)
~/anaconda3/envs/py3/lib/python3.7/site-packages/holoviews/core/data/__init__.py in pipelined_fn(*args, **kwargs)
214
215 try:
--> 216 result = method_fn(*args, **kwargs)
217
218 op = method_op.instance(
~/anaconda3/envs/py3/lib/python3.7/site-packages/holoviews/core/data/__init__.py in dimension_values(self, dimension, expanded, flat)
1069 """
1070 dim = self.get_dimension(dimension, strict=True)
-> 1071 return self.interface.values(self, dim, expanded, flat)
1072
1073
~/anaconda3/envs/py3/lib/python3.7/site-packages/holoviews/core/data/xarray.py in values(cls, dataset, dim, expanded, flat, compute, keep_index)
376 if not keep_index:
377 data = cls.canonicalize(dataset, data, data_coords=data_coords,
--> 378 virtual_coords=virtual_coords)
379 return data.T.flatten() if flat and not keep_index else data
380 elif expanded:
~/anaconda3/envs/py3/lib/python3.7/site-packages/holoviews/core/data/grid.py in canonicalize(cls, dataset, data, data_coords, virtual_coords)
341
342 if not any(cls.irregular(dataset, d) for d in dataset.kdims):
--> 343 inds = [dims.index(kd.name) for kd in dataset.kdims]
344 inds = [i - sum([1 for d in dropped if i>=d]) for i in inds]
345 if inds:
~/anaconda3/envs/py3/lib/python3.7/site-packages/holoviews/core/data/grid.py in <listcomp>(.0)
341
342 if not any(cls.irregular(dataset, d) for d in dataset.kdims):
--> 343 inds = [dims.index(kd.name) for kd in dataset.kdims]
344 inds = [i - sum([1 for d in dropped if i>=d]) for i in inds]
345 if inds:
ValueError: 'lon' is not in lis
:NdLayout [var,group]
:HoloMap [tau_day,season,region]
:Image [lon,lat] (values,var)
I've just done a major contribution of PerspectiveViewer, Tabulator and PivotTable widgets. Feel free to help out reviewing holoviz/panel#1690