@@RyanNoonan for example: create a layout with several plots, in which, you can choice some stock market and you can see its time serie and different metrics like performance and calculate correlation with another stock. Sorry for my English
Brother, your killing me :-). I watched this specifically to see how to plot points (lat/long) on a map. Then you pushed passed it in a second. My data is lat and long that I want to put on a map. Can you address that with holoview? I've done it with other tools but...I'm really liking this holoview... Thank you, and by the way, your tutorials are awesome!
You probably want to use GeoViews. GeoViews is built on the HoloViews library. Basically, GeoViews is HoloViews for maps. Here is an example: import pandas as pd import geoviews as gv gv.extension('bokeh') from vega_datasets import data as vds airports = vds.airports()[['name', 'city', 'state', 'country', 'latitude', 'longitude']][:15] data = gv.Dataset(airports) points = data.to(gv.Points, ['longitude', 'latitude'], ['city', 'state', 'name']) tiles = gv.tile_sources.Wikipedia tiles * points.opts(width=800, height=600, tools=['hover'], size=10)
@@RyanNoonan If I have a pandas data frame of lon, lat and a z field that i want color coated, what would that look like in geoviews? I have been trying to get this to work for days but it does not seem to like pandas or something? I am not sure. I can get it done in hvplot and mpl just fine. A whole video on geoviews would be great as i have not been able to find a single one. All the data in the examples are in xarrays or something and I have not once gotten real data that is in that format. A video on using it with normal data would be great.
I understand. This is how I think about it. There is some overlap between the packages. Bokeh is great for many types of data visualization - charts, maps, and some ability to create dashboards and has many options. HoloViews is similar to Bokeh but often you can visualize data with fewer lines of code. hvPlot is similar to HoloViews but often you can visualize data with even fewer lines of code and is really good when working with data frames to quickly generate plots. Panel is excellent for creating interactive dashboards.
@@RyanNoonan right, it would be nice to have "1 that rules them all". Regarding Holoviews and being a just starting to use this framework, I'm having issues with the documentation. It is not on the web page (or at least all the options) and checking the options from the console without any further explanation is a bit of a hustle. Thanks to your video I learned a few though!
Thanks for taking the time to make this tutorial. Definitely interested in more holoviews content please. Can the backend support plotly for plotting?
Thanks. Yes, I believe the backend can support plotly. What additional holoviews content would you be interested to see?
Perhaps, how about taking a dataset into Pandas and then plotting various charts / maps using Holoviews + Datashader? Thanks
Thanks! very helpful and knowledgeable!
I would like to know more about heatmaps (interpolated heatmaps specifically) and widgets
very interested in other visuals !!
That is one of the coolest videos!
One small query here, can we export the final plot in an excel sheet with those drop downs as well ?
Thanks. Interesting. If I find a way, I'll let you know.
Thanks 😉. It would be interesting introduction to holoviews + panel to created apps focus on data science.
I like that idea. Please share some specifics. Thanks.
@@RyanNoonan for example: create a layout with several plots, in which, you can choice some stock market and you can see its time serie and different metrics like performance and calculate correlation with another stock. Sorry for my English
Very nice introduction! thanks for your sharing!
Super cool..one video, all inclusive.
Can we work with maps/GeoJson in HoloView?
Thanks. Yes there are some map options but also check out the GeoViews package.
Any videos on Datashader + HoloViews? Specifically, using the curves and customising the tooltips, ranges, formatting..etc?
Not yet but it is on my list. Thanks for the suggestion.
Very nice overview - thanks a lot!
Brother, your killing me :-). I watched this specifically to see how to plot points (lat/long) on a map. Then you pushed passed it in a second. My data is lat and long that I want to put on a map. Can you address that with holoview? I've done it with other tools but...I'm really liking this holoview... Thank you, and by the way, your tutorials are awesome!
You probably want to use GeoViews. GeoViews is built on the HoloViews library. Basically, GeoViews is HoloViews for maps.
Here is an example:
import pandas as pd
import geoviews as gv
gv.extension('bokeh')
from vega_datasets import data as vds
airports = vds.airports()[['name', 'city', 'state', 'country', 'latitude', 'longitude']][:15]
data = gv.Dataset(airports)
points = data.to(gv.Points, ['longitude', 'latitude'], ['city', 'state', 'name'])
tiles = gv.tile_sources.Wikipedia
tiles * points.opts(width=800, height=600, tools=['hover'], size=10)
@@RyanNoonan Thank you. I'm trying now to work through it. I'm having issues installing cartopy but will figure that out. Thank you very much!
@@RyanNoonan If I have a pandas data frame of lon, lat and a z field that i want color coated, what would that look like in geoviews? I have been trying to get this to work for days but it does not seem to like pandas or something? I am not sure. I can get it done in hvplot and mpl just fine. A whole video on geoviews would be great as i have not been able to find a single one. All the data in the examples are in xarrays or something and I have not once gotten real data that is in that format. A video on using it with normal data would be great.
Thanks for the suggestion. At some point, I'll try to do a tutorial on geoviews and geoviews with data sources, etc.
Bokeh, Holoviews, Hvplot, Panel... I'm a little lost at the moment...
I understand. This is how I think about it. There is some overlap between the packages. Bokeh is great for many types of data visualization - charts, maps, and some ability to create dashboards and has many options. HoloViews is similar to Bokeh but often you can visualize data with fewer lines of code. hvPlot is similar to HoloViews but often you can visualize data with even fewer lines of code and is really good when working with data frames to quickly generate plots. Panel is excellent for creating interactive dashboards.
@@RyanNoonan right, it would be nice to have "1 that rules them all". Regarding Holoviews and being a just starting to use this framework, I'm having issues with the documentation. It is not on the web page (or at least all the options) and checking the options from the console without any further explanation is a bit of a hustle. Thanks to your video I learned a few though!