Add labels to points to aid data exploration

Development matplotlib-devel
1

I posted this to https://github.com/matplotlib/matplotlib/issues/2856
but it was suggested that the mailing list was a better place.

I am plotting a simple graph with many points. I would like to explore
the data by clicking on some of them to see what their labels are
(much as I can already in R). I don't want to see all the labels at
once. Currently the only solution I found was mpldatacursor. Is this
feature something interesting to matplotlib developers or even being
currently worked on?

Raphael

2

Hey Raphael,

Jake Vanderplas has been working on mpld3, which lets you create D3 renderings of matplotlib graphs. We (Plotly) are working with him on a matplotlib exporter that lets you make interactive, online, labels/text on the hover Plotly graphs with matplotlib. Here is an IPython Notebook showing how to do that:

http://nbviewer.ipython.org/github/theengineear/mplexporter/blob/master/notebooks/PlotlyTest.ipynb?create=1

We’re early on still, and Jake is definitely the driving force/mastermind behind the project. We’re hoping to support more and more plotting from matplotlib and would love help, tips, feedback, and suggestions. We’re happy to talk more if you’d like.

Thanks a bunch,

M

···

On Sun, Mar 2, 2014 at 12:17 PM, Raphael C <drraph@…149…> wrote:

I posted this to https://github.com/matplotlib/matplotlib/issues/2856

but it was suggested that the mailing list was a better place.

I am plotting a simple graph with many points. I would like to explore

the data by clicking on some of them to see what their labels are

(much as I can already in R). I don’t want to see all the labels at

once. Currently the only solution I found was mpldatacursor. Is this

feature something interesting to matplotlib developers or even being

currently worked on?

Raphael

Flow-based real-time traffic analytics software. Cisco certified tool.

Monitor traffic, SLAs, QoS, Medianet, WAAS etc. with NetFlow Analyzer

Customize your own dashboards, set traffic alerts and generate reports.

Network behavioral analysis & security monitoring. All-in-one tool.

http://pubads.g.doubleclick.net/gampad/clk?id=126839071&iu=/4140/ostg.clktrk

Matplotlib-devel mailing list

Matplotlib-devel@lists.sourceforge.net

https://lists.sourceforge.net/lists/listinfo/matplotlib-devel

3

Thank you that is good to know. From my personal point of view I was
just interested in a view simple form of interactivity. That is the
ability to click on data points and see their labels. I have to say
mpldatacursor is very nice and does precisely this. It however has
some disadvantages.

a) It's not possible to save the current state of your interactive
visualisation to send to someone else (except as a snapshot png).
b) Due to its dependencies it is also hard to send your code to
someone else as they are unlikely to have the dependencies to get it
to work. This would be solved by having the functionality in something
standard like matplotlib.
c) There appears to be only one contributor currently. (I could be
wrong about this last point.)

Raphael

···

On 2 March 2014 20:32, Matt Sundquist <matt@...1186...> wrote:

Hey Raphael,

Jake Vanderplas has been working on mpld3, which lets you create D3
renderings of matplotlib graphs. We (Plotly) are working with him on a
matplotlib exporter that lets you make interactive, online, labels/text on
the hover Plotly graphs with matplotlib. Here is an IPython Notebook showing
how to do that:

http://nbviewer.ipython.org/github/theengineear/mplexporter/blob/master/notebooks/PlotlyTest.ipynb?create=1

We're early on still, and Jake is definitely the driving force/mastermind
behind the project. We're hoping to support more and more plotting from
matplotlib and would love help, tips, feedback, and suggestions. We're happy
to talk more if you'd like.

Thanks a bunch,
M

On Sun, Mar 2, 2014 at 12:17 PM, Raphael C <drraph@...149...> wrote:

I posted this to Add labels to points to aid data exploration · Issue #2856 · matplotlib/matplotlib · GitHub
but it was suggested that the mailing list was a better place.

I am plotting a simple graph with many points. I would like to explore
the data by clicking on some of them to see what their labels are
(much as I can already in R). I don't want to see all the labels at
once. Currently the only solution I found was mpldatacursor. Is this
feature something interesting to matplotlib developers or even being
currently worked on?

Raphael

------------------------------------------------------------------------------
Flow-based real-time traffic analytics software. Cisco certified tool.
Monitor traffic, SLAs, QoS, Medianet, WAAS etc. with NetFlow Analyzer
Customize your own dashboards, set traffic alerts and generate reports.
Network behavioral analysis & security monitoring. All-in-one tool.

http://pubads.g.doubleclick.net/gampad/clk?id=126839071&iu=/4140/ostg.clktrk
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
matplotlib-devel List Signup and Options

4

Is this something the new WebAgg backend should solve?

Raphael

···

On 2 March 2014 20:46, Raphael C <drraph@...149...> wrote:

Thank you that is good to know. From my personal point of view I was
just interested in a view simple form of interactivity. That is the
ability to click on data points and see their labels. I have to say
mpldatacursor is very nice and does precisely this. It however has
some disadvantages.

a) It's not possible to save the current state of your interactive
visualisation to send to someone else (except as a snapshot png).
b) Due to its dependencies it is also hard to send your code to
someone else as they are unlikely to have the dependencies to get it
to work. This would be solved by having the functionality in something
standard like matplotlib.
c) There appears to be only one contributor currently. (I could be
wrong about this last point.)

Raphael

On 2 March 2014 20:32, Matt Sundquist <matt@...1186...> wrote:

Hey Raphael,

Jake Vanderplas has been working on mpld3, which lets you create D3
renderings of matplotlib graphs. We (Plotly) are working with him on a
matplotlib exporter that lets you make interactive, online, labels/text on
the hover Plotly graphs with matplotlib. Here is an IPython Notebook showing
how to do that:

http://nbviewer.ipython.org/github/theengineear/mplexporter/blob/master/notebooks/PlotlyTest.ipynb?create=1

We're early on still, and Jake is definitely the driving force/mastermind
behind the project. We're hoping to support more and more plotting from
matplotlib and would love help, tips, feedback, and suggestions. We're happy
to talk more if you'd like.

Thanks a bunch,
M

On Sun, Mar 2, 2014 at 12:17 PM, Raphael C <drraph@...149...> wrote:

I posted this to Add labels to points to aid data exploration · Issue #2856 · matplotlib/matplotlib · GitHub
but it was suggested that the mailing list was a better place.

I am plotting a simple graph with many points. I would like to explore
the data by clicking on some of them to see what their labels are
(much as I can already in R). I don't want to see all the labels at
once. Currently the only solution I found was mpldatacursor. Is this
feature something interesting to matplotlib developers or even being
currently worked on?

Raphael

------------------------------------------------------------------------------
Flow-based real-time traffic analytics software. Cisco certified tool.
Monitor traffic, SLAs, QoS, Medianet, WAAS etc. with NetFlow Analyzer
Customize your own dashboards, set traffic alerts and generate reports.
Network behavioral analysis & security monitoring. All-in-one tool.

http://pubads.g.doubleclick.net/gampad/clk?id=126839071&iu=/4140/ostg.clktrk
_______________________________________________
Matplotlib-devel mailing list
Matplotlib-devel@lists.sourceforge.net
matplotlib-devel List Signup and Options