[TriPython] Prediction Model. Data Visualization.
Francois Dion
francois.dion at gmail.com
Tue Oct 10 21:41:57 EDT 2017
Without the images, it is a bit hard to see what you are trying to achieve.
If what you are trying to do is represent a confusion matrix with a semi
graphical output (i'm all for that), then have a look at Yellow Brick. The
confusion matrix is here:
http://www.scikit-yb.org/en/latest/api/classifier/confusion_matrix.html
They also have a few more visualizations for classification:
http://www.scikit-yb.org/en/latest/#classification-visualization
Seaborn has a heatmap, so that can be used for a CM too. Now, if it has to
be on the web, interactive, look at bokeh or plot.ly. I did a plotly
presentation at PYPTUG last month, some code and a lot more stuff can be
found here: https://github.com/fdion/pyptug_plotly
Francois
On Tue, Oct 10, 2017 at 3:53 PM, Art <artem.nesterenko at gmail.com> wrote:
> Good afternoon!
> I'm reaching out to you guys for a suggestion on the data visualization.
> I'm wondering if anyone has an experience or an idea of visualizing the
> volume of multiple model prediction targets.
> Here is an example of a model with just 2 prediction targets I've
> recently
> been working on:
> 1. This model provides 2 predictions: target_1 and target_2.**
> 2. I also know the actual result to compare against, so I could see the
> model accuracy.**
> 3. I've built a confusion matrix to calculate true/false pos/neg (see
> below). So, there are 4 values.
> 4. I chose donut**chart to visualize these values (see below).
> [1]Inline image 2**
>
> My question is what if the model provides more than 2 predictions? Our
> next model has 7 targets, which is 49 pos and neg values as far as I
> understand. And I think a**donut**graph is not going to work in this
> case.**Maybe a bar graph or something else fits better.**
>
> I'd appreciate any ideas or examples of visualizations that easy to look
> at and understand.
> I'm using d3/c3.js for data visualization.
> Thank you!
> Art Nestsiarenka
> email: [2]artem.nesterenko at gmail.com
>
> References
>
> Visible links
> 2. mailto:artem.nesterenko at gmail.com
>
> _______________________________________________
> TriZPUG mailing list
> TriZPUG at python.org
> https://mail.python.org/mailman/listinfo/trizpug
> http://tripython.org is the Triangle Python Users Group
>
>
--
about.me/francois.dion - www.pyptug.org - www.3DFutureTech.info - @f_dion
<http://twitter.com/f_dion>
-------------- next part --------------
Without the images, it is a bit hard to see what you are trying to
achieve.
If what you are trying to do is represent a confusion matrix with a semi
graphical output (i'm all for that), then have a look at Yellow Brick. The
confusion matrix is here:
[1]http://www.scikit-yb.org/en/latest/api/classifier/confusion_matrix.html
They also have a few more visualizations for classification:
[2]http://www.scikit-yb.org/en/latest/#classification-visualization
Seaborn has a heatmap, so that can be used for a CM too. Now, if it has to
be on the web, interactive, look at bokeh or [3]plot.ly. I did a plotly
presentation at PYPTUG last month, some code and a lot more stuff can be
found here: [4]https://github.com/fdion/pyptug_plotly
Francois
On Tue, Oct 10, 2017 at 3:53 PM, Art <[5]artem.nesterenko at gmail.com>
wrote:
** **Good afternoon!
** **I'm reaching out to you guys for a suggestion on the data
visualization.
** **I'm wondering if anyone has an experience or an idea of visualizing
the
** **volume of multiple model prediction targets.
** **Here is an example of a model with just 2 prediction targets I've
recently
** **been working on:
** **1. This model provides 2 predictions: target_1 and target_2.**
** **2. I also know the actual result to compare against, so I could see
the
** **model accuracy.**
** **3. I've built a confusion matrix to calculate true/false pos/neg
(see
** **below). So, there are 4 values.
** **4. I chose donut**chart to visualize these values (see below).
** **[1]Inline image 2**
** **My question is what if the model provides more than 2 predictions?
Our
** **next model has 7 targets, which is 49 pos and neg values as far as
I
** **understand. And I think a**donut**graph is not going to work in
this
** **case.**Maybe a bar graph or something else fits better.**
** **I'd appreciate any ideas or examples of visualizations that easy to
look
** **at and understand.
** **I'm using d3/c3.js for data visualization.
** **Thank you!
** **Art Nestsiarenka
** **email: [2][6]artem.nesterenko at gmail.com
References
** **Visible links
** **2. mailto:[7]artem.nesterenko at gmail.com
_______________________________________________
TriZPUG mailing list
[8]TriZPUG at python.org
[9]https://mail.python.org/mailman/listinfo/trizpug
[10]http://tripython.org is the Triangle Python Users Group
--
[11]about.me/francois.dion - [12]www.pyptug.org -
[13]www.3DFutureTech.info - [14]@f_dion
References
Visible links
1. http://www.scikit-yb.org/en/latest/api/classifier/confusion_matrix.html
2. http://www.scikit-yb.org/en/latest/#classification-visualization
3. http://plot.ly/
4. https://github.com/fdion/pyptug_plotly
5. mailto:artem.nesterenko at gmail.com
6. mailto:artem.nesterenko at gmail.com
7. mailto:artem.nesterenko at gmail.com
8. mailto:TriZPUG at python.org
9. https://mail.python.org/mailman/listinfo/trizpug
10. http://tripython.org/
11. http://about.me/francois.dion
12. http://www.pyptug.org/
13. http://www.3dfuturetech.info/
14. http://twitter.com/f_dion
More information about the TriZPUG
mailing list